Trong bối cảnh khan hiếm nước ngọt ngày càng trầm trọng trên toàn cầu, công nghệ trí tuệ nhân tạo (AI) đang nổi lên như một giải pháp đột phá để giảm thiểu lãng phí nguồn tài nguyên quý giá này. Chủ đề “AI For Reducing Water Waste” không chỉ mang tính thời sự cao mà còn thường xuyên xuất hiện trong các kỳ thi IELTS Reading, đặc biệt ở dạng bài về khoa học công nghệ và môi trường.
Bài viết này cung cấp cho bạn một bộ đề thi IELTS Reading hoàn chỉnh với 3 passages từ dễ đến khó, bao gồm 40 câu hỏi đa dạng theo đúng format thi thật. Bạn sẽ được luyện tập với các dạng câu hỏi phổ biến như Multiple Choice, True/False/Not Given, Matching Headings, Summary Completion và nhiều dạng khác. Mỗi câu hỏi đều có đáp án chi tiết kèm giải thích, giúp bạn hiểu rõ phương pháp làm bài và cách paraphrase của đề thi.
Đề thi này phù hợp cho học viên từ band 5.0 trở lên, với độ khó tăng dần giúp bạn làm quen với áp lực thời gian và yêu cầu kỹ năng đọc hiểu ở các mức độ khác nhau. Hãy chuẩn bị đồng hồ bấm giờ, tạo môi trường thi thật và bắt đầu thử thách 60 phút của bạn!
Hướng Dẫn Làm Bài IELTS Reading
Tổng Quan Về IELTS Reading Test
IELTS Reading Test là phần thi kéo dài 60 phút với 3 passages và tổng cộng 40 câu hỏi. Mỗi câu trả lời đúng được tính 1 điểm, không bị trừ điểm khi sai. Điểm số sau đó được quy đổi thành band điểm từ 1-9.
Phân bổ thời gian khuyến nghị:
- Passage 1: 15-17 phút (độ khó thấp, câu hỏi tương đối dễ tìm đáp án)
- Passage 2: 18-20 phút (độ khó trung bình, yêu cầu kỹ năng paraphrase tốt)
- Passage 3: 23-25 phút (độ khó cao, nội dung phức tạp, cần phân tích sâu)
Lưu ý dành 2-3 phút cuối để chuyển đáp án vào answer sheet và kiểm tra lại.
Các Dạng Câu Hỏi Trong Đề Này
Đề thi mẫu này bao gồm 8 dạng câu hỏi phổ biến nhất trong IELTS Reading:
- Multiple Choice – Chọn đáp án đúng từ các phương án cho sẵn
- True/False/Not Given – Xác định thông tin đúng, sai hoặc không được đề cập
- Matching Information – Nối thông tin với đoạn văn tương ứng
- Sentence Completion – Hoàn thành câu với thông tin từ bài đọc
- Matching Headings – Chọn tiêu đề phù hợp cho mỗi đoạn
- Summary Completion – Điền từ vào đoạn tóm tắt
- Matching Features – Nối đặc điểm với danh mục cho sẵn
- Short-answer Questions – Trả lời câu hỏi ngắn với từ trong bài
IELTS Reading Practice Test
PASSAGE 1 – Smart Water Management in Urban Agriculture
Độ khó: Easy (Band 5.0-6.5)
Thời gian đề xuất: 15-17 phút
Urban agriculture is experiencing a technological revolution, with artificial intelligence (AI) playing a central role in addressing one of the sector’s most critical challenges: water waste. As cities expand and fresh water becomes increasingly scarce, farmers and gardeners in urban environments are turning to intelligent systems to optimize their water usage and ensure sustainable food production.
In traditional farming methods, irrigation typically follows fixed schedules, with water applied at predetermined times regardless of actual plant needs or weather conditions. This approach often results in overwatering, which not only wastes precious water resources but can also damage crops by creating waterlogged soil conditions and encouraging root diseases. Conversely, some areas may receive insufficient water, leading to stunted growth and reduced yields. The problem is particularly acute in urban settings where space is limited and every drop of water counts.
AI-powered irrigation systems are changing this landscape dramatically. These systems use sensors placed throughout growing areas to continuously monitor multiple factors including soil moisture levels, temperature, humidity, and even plant health indicators. The collected data is transmitted to a central AI processor that analyzes the information in real-time. By comparing current conditions with historical data and weather forecasts, the AI can determine precisely when and how much water each section of crops requires.
One of the most impressive features of these systems is their ability to learn and adapt. Machine learning algorithms improve their accuracy over time by observing the relationships between watering patterns, plant growth, and environmental conditions. For instance, if the system notices that certain plants consistently show better growth with slightly less water on sunny days following rain, it will automatically adjust future watering schedules accordingly. This adaptive capability ensures that the system becomes more efficient the longer it operates.
The impact of AI on water conservation in urban agriculture has been remarkable. Studies conducted in several major cities have shown that AI-managed irrigation systems can reduce water consumption by 30 to 50 percent compared to traditional methods, while simultaneously improving crop yields by 15 to 25 percent. In Tokyo, a rooftop farming initiative equipped with AI irrigation reported saving over 2 million liters of water annually while producing 20 percent more vegetables than comparable farms using conventional watering methods.
The technology is also becoming more accessible to small-scale urban farmers and home gardeners. Several companies now offer affordable AI irrigation solutions that can be installed without professional help. These systems connect to smartphone apps, allowing users to monitor their gardens remotely and receive alerts about watering needs or potential problems. Some advanced versions even integrate with local weather services to automatically adjust watering plans when rain is forecast, ensuring no water is wasted on irrigation just before a natural rainfall.
Beyond individual farms, cities are implementing AI water management at a municipal level. Singapore’s “Smart Nation” initiative includes AI systems that manage water distribution across all urban agricultural spaces, from community gardens to commercial vertical farms. The system detects leaks in irrigation infrastructure, predicts maintenance needs, and redistributes water resources to areas of greatest need. This coordinated approach has helped the city-state reduce agricultural water waste by 40 percent since 2018.
Despite these successes, challenges remain. The initial cost of AI systems, while decreasing, can still be prohibitive for some small-scale farmers. There are also concerns about data privacy and the need for reliable internet connectivity. Additionally, some traditional farmers are skeptical about relying on technology rather than their own experience and judgment. However, as success stories multiply and costs continue to fall, adoption rates are steadily increasing.
Looking ahead, researchers are developing even more sophisticated AI applications for water management. Predictive models are being created that can forecast plant water needs days in advance based on growth stages and anticipated weather patterns. Some systems are experimenting with drone technology to provide aerial monitoring of large urban farms, identifying areas of water stress before they become visible to the human eye. Others are exploring the integration of AI with rainwater harvesting systems, automatically switching between harvested rainwater and municipal supplies to maximize the use of free natural resources.
Questions 1-13
Questions 1-5
Do the following statements agree with the information given in the passage?
Write:
- TRUE if the statement agrees with the information
- FALSE if the statement contradicts the information
- NOT GIVEN if there is no information on this
- Traditional irrigation methods apply water according to actual plant requirements.
- Overwatering can cause root diseases in crops.
- AI irrigation systems use sensors to monitor soil moisture and temperature.
- Machine learning algorithms in AI systems improve their performance over time.
- All small-scale urban farmers have adopted AI irrigation technology.
Questions 6-9
Complete the sentences below.
Choose NO MORE THAN TWO WORDS AND/OR A NUMBER from the passage for each answer.
- AI-managed irrigation systems can reduce water use by up to __ compared to traditional methods.
- A rooftop farm in Tokyo saved over __ of water each year using AI irrigation.
- Some AI systems connect to __ to notify users about garden conditions.
- Singapore reduced agricultural water waste by __ since implementing its Smart Nation initiative.
Questions 10-13
Choose the correct letter, A, B, C or D.
-
According to the passage, what is a main problem with fixed-schedule irrigation?
- A) It is too expensive to maintain
- B) It does not consider actual plant needs
- C) It requires too much manual labor
- D) It only works in large farms
-
What happens when the AI system notices patterns in plant growth?
- A) It sends a report to farmers
- B) It automatically adjusts future watering
- C) It stops watering completely
- D) It increases water usage
-
What is mentioned as a barrier to AI adoption?
- A) Lack of government support
- B) Complicated installation process
- C) High initial costs
- D) Limited crop varieties
-
Future AI systems are being developed to:
- A) Replace all human farmers
- B) Work without internet
- C) Forecast plant water needs in advance
- D) Reduce crop yields
PASSAGE 2 – Industrial Applications of AI in Water Loss Prevention
Độ khó: Medium (Band 6.0-7.5)
Thời gian đề xuất: 18-20 phút
Water scarcity represents one of the most pressing challenges facing modern civilization, with the United Nations estimating that by 2025, half of the world’s population will be living in water-stressed regions. While much attention focuses on consumption reduction, an equally critical issue receives less public attention: the enormous volumes of water lost through leaks, inefficiencies, and system failures in industrial and municipal water infrastructure. Here, artificial intelligence is emerging as a game-changing technology, offering unprecedented capabilities to detect, predict, and prevent water loss on a massive scale.
A. The Scale of the Problem
The statistics surrounding water waste in distribution systems are staggering. In developed nations, an estimated 20 to 30 percent of treated water never reaches its destination, lost instead through aging pipes, faulty joints, and undetected leaks. The situation in developing countries is often far worse, with some cities losing more than 50 percent of their water supply before it reaches consumers. In purely economic terms, this represents billions of dollars in wasted resources annually. From an environmental perspective, it means unnecessary strain on water sources and the energy-intensive processes required for water treatment and pumping are being expended on water that ultimately serves no purpose.
B. Traditional Detection Methods and Their Limitations
Historically, water utilities have relied on manual inspection and pressure monitoring to identify leaks. Teams of workers would physically patrol pipeline networks, listening for the sound of escaping water or looking for visible signs such as puddles or unusual vegetation growth. While this approach can identify major ruptures, it is woefully inadequate for detecting small leaks, which collectively account for the majority of water loss. Acoustic sensors represented an improvement, allowing technicians to identify leak sounds in pipes, but their effectiveness is limited by background noise in urban environments and the need for time-consuming manual deployment across vast network areas.
C. AI-Powered Leak Detection Systems
Modern AI solutions employ a fundamentally different approach. Rather than searching for individual leaks, these systems continuously analyze data streams from thousands of sensors distributed throughout water networks. These sensors monitor pressure, flow rates, water quality, and even acoustic signatures at numerous points. The AI algorithms process this massive dataset in real-time, searching for anomalous patterns that indicate water loss. A sudden pressure drop in a specific pipe section, unusual flow patterns during low-demand periods, or acoustic frequencies consistent with water escaping under pressure can all trigger alerts for investigation.
What makes AI particularly powerful in this context is its ability to detect subtle changes that would be imperceptible to human observers. A small leak might cause only a negligible pressure change at any single monitoring point, but by analyzing data from multiple sensors simultaneously, AI can identify the cumulative effect and triangulate the leak’s location with remarkable precision. One system deployed in Barcelona, Spain, managed to pinpoint leaks to within three meters in a network spanning hundreds of kilometers, reducing the time and cost of repairs dramatically.
D. Predictive Maintenance and Failure Prevention
Perhaps even more significant than leak detection is AI’s capacity for predictive maintenance. By analyzing historical failure data alongside current operational parameters, AI systems can identify pipes and components at high risk of failure before they actually break. Factors such as pipe age, material composition, soil conditions, pressure fluctuations, and historical weather patterns all contribute to failure probability. The AI evaluates these variables to generate risk assessments for different network sections, allowing utilities to prioritize proactive repairs and replacements.
The city of Doha, Qatar, implemented such a system across its water network in 2019. The AI analyzed data from 50,000 sensors along with records of past failures spanning two decades. Within the first year of operation, the system successfully predicted 76 percent of major failures at least two weeks in advance, allowing repair crews to address issues during scheduled maintenance rather than responding to emergencies. This shift from reactive to proactive management reduced water loss by 28 percent and cut emergency repair costs by $12 million annually.
E. Industrial Manufacturing Applications
Beyond municipal water systems, AI is transforming water management in manufacturing industries, where water serves critical roles in cooling, cleaning, and production processes. Many industrial facilities consume millions of liters daily, and even minor inefficiencies or leaks represent substantial waste. AI systems monitor industrial water circuits, identifying optimization opportunities and detecting losses that traditional methods would miss.
A semiconductor manufacturing plant in Taiwan provides a compelling case study. Semiconductor production requires ultra-pure water in vast quantities, with a single modern facility consuming up to 20 million liters per day. The plant implemented an AI monitoring system that tracks water usage across hundreds of individual processes. By identifying inefficient rinse cycles, detecting micro-leaks in cooling systems, and optimizing water recycling processes, the AI helped reduce overall water consumption by 18 percent—equivalent to saving over 1.3 billion liters annually. The system paid for itself within eight months through reduced water and energy costs.
F. Integration Challenges and Future Directions
Despite these impressive results, widespread AI adoption in water management faces several obstacles. Many water utilities operate with legacy infrastructure and limited budgets, making the transition to smart systems financially challenging. There are also interoperability issues when attempting to integrate AI platforms with existing monitoring equipment from different manufacturers. Cybersecurity concerns represent another significant consideration, as networked water systems could potentially become targets for malicious attacks.
Looking forward, researchers are developing next-generation AI systems that require fewer sensors by using satellite imagery and ground-penetrating radar data to identify potential problem areas. Other innovations include AI that can optimize entire watershed management systems, considering factors from rainfall prediction to consumption patterns to minimize waste across entire regions rather than just individual networks.
Hệ thống AI quản lý nước thông minh với cảm biến và phân tích dữ liệu thời gian thực cho ngành công nghiệp
Questions 14-26
Questions 14-18
The passage has six sections, A-F.
Which section contains the following information?
Write the correct letter, A-F.
- Examples of factors analyzed to predict pipe failures
- Statistics about water loss in global distribution systems
- A description of how traditional leak detection methods work
- Information about water usage in semiconductor production
- Mention of security risks associated with networked systems
Questions 19-22
Complete the summary below.
Choose NO MORE THAN TWO WORDS from the passage for each answer.
AI systems for water management use thousands of sensors to monitor various parameters including pressure and (19) __. These systems can detect (20) ____ that indicate water loss. Unlike human observers, AI can identify (21) ____ that are too small to notice individually. In Barcelona, one system could locate leaks to within (22) ____ in a vast network.
Questions 23-26
Choose the correct letter, A, B, C or D.
-
According to the passage, what percentage of water is typically lost in developed nations?
- A) 10 to 15 percent
- B) 20 to 30 percent
- C) More than 50 percent
- D) 75 percent
-
What was the key achievement of Doha’s AI water system?
- A) It eliminated all water leaks
- B) It predicted most major failures in advance
- C) It reduced sensor requirements
- D) It replaced all manual workers
-
How quickly did the Taiwan semiconductor plant’s AI system pay for itself?
- A) Within eight months
- B) Within one year
- C) Within two years
- D) Within five years
-
What is mentioned as a future development in water management AI?
- A) Completely autonomous systems
- B) Use of satellite imagery to reduce sensor needs
- C) Replacement of all pipes
- D) Elimination of all water waste
PASSAGE 3 – The Algorithmic Optimization of Global Water Resources: Complexities and Controversies
Độ khó: Hard (Band 7.0-9.0)
Thời gian đề xuất: 23-25 phút
The advent of sophisticated artificial intelligence systems capable of analyzing and optimizing water usage across entire regions represents a paradigm shift in resource management, yet this technological advancement exists at the intersection of numerous complex ethical, social, and technical considerations that demand careful examination. While the potential for AI to substantially reduce water waste is indisputable, the implementation of such systems raises fundamental questions about governance, equity, and the nature of algorithmic decision-making in contexts where access to water constitutes a basic human right and an essential prerequisite for survival.
The technical architecture underpinning regional water management AI systems operates at a scale and complexity that challenges human comprehension. These systems integrate disparate data streams from meteorological sensors, satellite-based Earth observation platforms, groundwater monitoring networks, municipal consumption records, agricultural usage patterns, and industrial demand forecasts. Through deep learning neural networks trained on decades of historical data, the AI constructs multidimensional models of water flow through both natural and artificial systems, identifying optimization opportunities invisible to conventional analysis. The algorithms can, for instance, predict with remarkable accuracy how altering reservoir release schedules might affect downstream agricultural productivity, urban supply reliability, and ecosystem health simultaneously—a multivariate optimization problem that exceeds the capacity of traditional planning methodologies.
However, optimization itself proves to be a conceptually problematic objective when applied to resources with profound social significance. The mathematical frameworks employed by AI systems require explicit objective functions—quantifiable goals the algorithm attempts to maximize or minimize. In contexts such as financial trading or manufacturing logistics, defining these objectives is relatively straightforward: maximize profit, minimize cost, optimize throughput. Water management presents far more intractable challenges. Should the system prioritize economic efficiency, directing water toward uses that generate the greatest GDP contribution? Should it emphasize egalitarian distribution, ensuring equal per capita access regardless of economic considerations? Should environmental preservation take precedence, maintaining ecosystem water flows even at the cost of human consumption? These questions lack purely technical answers and instead reflect value judgments that different societies and stakeholders answer divergently.
The case of the Murray-Darling Basin in Australia illustrates these tensions vividly. This massive river system, which sustains agriculture producing approximately 40 percent of Australia’s food supply while supporting internationally significant wetlands, has been subject to increasingly contentious water allocation debates for decades. In 2020, authorities implemented an AI-assisted management system designed to optimize water distribution across competing demands. The algorithm’s recommendations, derived from econometric modeling of agricultural productivity and hydrological simulation of environmental flows, frequently diverged sharply from allocations determined through the basin’s existing stakeholder consultation processes. Critics argued that the AI system, while technically sophisticated, essentially commodified water, reducing complex social and ecological relationships to computational abstractions. The controversy highlighted a fundamental tension: AI systems excel at optimization within defined parameters but cannot independently determine what should be optimized or resolve conflicts between incommensurable values.
Furthermore, the implementation of AI water management systems carries significant implications for power dynamics and decision-making authority. Traditional water governance typically involves deliberative processes engaging diverse stakeholders—farmers, environmental groups, indigenous communities, urban authorities, and industrial users—each with legitimate interests and local knowledge. The introduction of AI-generated recommendations risks creating a technocratic framework wherein complex social negotiations are superseded by algorithmic outputs presented as objective and scientific, despite their embededness in particular modeling assumptions and value priorities. This phenomenon, which scholars term “algorithmic authority,” can effectively disenfranchise stakeholders who lack technical expertise to critique the AI’s methodology or challenge its recommendations on equal epistemic footing.
The issue becomes particularly acute when considering historical inequities in water access. In many regions, current water distribution patterns reflect legacy effects of colonialism, discriminatory policies, and unequal power relationships. An AI system optimizing from current baseline conditions may inadvertently perpetuate these inequities, as algorithmic optimization tends toward incremental improvement rather than structural transformation. Research by the Institute for Water Justice documented cases where AI-recommended allocations in several developing nations systematically disadvantaged smallholder farmers and marginalized communities in favor of larger agricultural enterprises, essentially encoding existing disparities into ostensibly neutral technical solutions.
The technical limitations of current AI systems compound these concerns. Machine learning algorithms demonstrate well-documented vulnerabilities to unexpected conditions, performing poorly when confronted with situations substantially different from their training data. In the context of water management, climate change is generating precisely such novel conditions—precipitation patterns, temperature extremes, and hydrological behaviors increasingly divergent from historical norms. An AI trained on twentieth-century data may provide suboptimal or even counterproductive recommendations when faced with twenty-first-century climate realities. The 2021 collapse of a AI-managed irrigation scheduling system during unprecedented heat waves in the Pacific Northwest of North America demonstrated this vulnerability dramatically, as the algorithm failed to account for soil moisture dynamics under temperature conditions absent from its training dataset.
Moreover, the centralization of water management through AI systems creates potential vulnerabilities to catastrophic failure. Distributed decision-making systems, while perhaps less optimized, possess inherent redundancy and resilience. If individual communities or utilities manage their resources imperfectly but independently, failures remain localized. Conversely, a regional AI system, despite superior performance under normal conditions, represents a single point of failure. Technical malfunctions, cyber attacks, or flawed algorithmic logic could cascade across entire regions. The 2023 cyber incident affecting South Africa’s national water grid, wherein hackers manipulated AI control parameters to create artificial scarcity conditions in certain districts, provided a sobering illustration of these risks.
Despite these substantial concerns, dismissing AI’s potential contribution to water conservation would be equally problematic given the magnitude of current waste and the urgency of water scarcity challenges. The path forward likely involves hybrid governance models that leverage AI’s analytical capabilities while maintaining human oversight and democratic accountability. Some jurisdictions are experimenting with “AI advisory systems” that generate recommendations for human decision-makers rather than autonomous control, preserving space for stakeholder input and contextual judgment. Others are developing participatory AI design processes wherein affected communities contribute to defining the system’s objectives and constraints, potentially addressing concerns about algorithmic authority and encoded bias.
The European Union’s Water Framework Directive Revision, currently under consideration, proposes mandatory algorithmic transparency requirements for any AI systems influencing water allocation decisions. Under this framework, the systems would need to provide explainable outputs detailing the reasoning behind recommendations, allowing stakeholders to understand and potentially challenge the algorithmic logic. While technical challenges to explainability in complex neural networks remain significant, such regulatory approaches represent efforts to balance technological innovation with democratic governance principles.
Questions 27-40
Questions 27-31
Choose the correct letter, A, B, C or D.
-
According to the passage, what makes defining objectives for water management AI particularly difficult?
- A) The lack of available data
- B) The complexity of the software
- C) The involvement of value judgments
- D) The cost of implementation
-
The Murray-Darling Basin case study demonstrates that:
- A) AI systems always produce better results than human decision-makers
- B) Technical sophistication does not resolve conflicts between different values
- C) Agricultural productivity should be prioritized over environmental concerns
- D) Stakeholder consultation is unnecessary when using AI
-
What does the term “algorithmic authority” refer to?
- A) The legal power of AI systems
- B) The technical superiority of algorithms
- C) The way AI outputs are treated as objective and difficult to challenge
- D) The ability of algorithms to make final decisions
-
According to the passage, what problem can arise when AI optimizes from current conditions?
- A) It uses too much computational power
- B) It may perpetuate existing inequities
- C) It always favors environmental concerns
- D) It requires excessive stakeholder input
-
The Pacific Northwest irrigation system failure showed that:
- A) AI systems work perfectly in all conditions
- B) Climate change is not affecting water management
- C) AI trained on historical data may fail under novel conditions
- D) Temperature has no effect on irrigation needs
Questions 32-36
Complete the sentences below.
Choose NO MORE THAN THREE WORDS from the passage for each answer.
- AI water management systems integrate data from sources including meteorological sensors and __.
- The algorithms can predict how changing reservoir schedules affects agricultural productivity and __.
- Critics of AI water systems argue that they reduce complex relationships to __.
- Research showed AI recommendations in developing nations sometimes disadvantaged __.
- The 2023 South Africa incident involved hackers manipulating AI to create artificial __.
Questions 37-40
Do the following statements agree with the claims of the writer in the passage?
Write:
- YES if the statement agrees with the claims of the writer
- NO if the statement contradicts the claims of the writer
- NOT GIVEN if it is impossible to say what the writer thinks about this
- AI systems for water management are technically superior to all traditional methods in every situation.
- Centralized AI water management systems may be vulnerable to catastrophic failures.
- Hybrid governance models could combine AI capabilities with human oversight.
- The European Union has already implemented mandatory transparency requirements for water management AI.
Mô hình AI quản lý nguồn nước quốc gia với hệ thống cảm biến phân tán và trung tâm dữ liệu
Answer Keys – Đáp Án
PASSAGE 1: Questions 1-13
- FALSE
- TRUE
- TRUE
- TRUE
- NOT GIVEN
- 50 percent / 50%
- 2 million liters
- smartphone apps
- 40 percent / 40%
- B
- B
- C
- C
PASSAGE 2: Questions 14-26
- D
- A
- B
- E
- F
- flow rates
- anomalous patterns
- subtle changes
- three meters
- B
- B
- A
- B
PASSAGE 3: Questions 27-40
- C
- B
- C
- B
- C
- satellite-based Earth observation platforms (hoặc: Earth observation platforms)
- ecosystem health
- computational abstractions
- smallholder farmers
- scarcity conditions
- NO
- YES
- YES
- NOT GIVEN
Giải Thích Đáp Án Chi Tiết
Passage 1 – Giải Thích
Câu 1: FALSE
- Dạng câu hỏi: True/False/Not Given
- Từ khóa: traditional irrigation methods, actual plant requirements
- Vị trí trong bài: Đoạn 2, dòng 1-2
- Giải thích: Bài đọc nói rõ “irrigation typically follows fixed schedules, with water applied at predetermined times regardless of actual plant needs” – tức là KHÔNG theo nhu cầu thực tế của cây. Câu hỏi nói phương pháp truyền thống áp dụng nước theo yêu cầu thực tế của cây, điều này mâu thuẫn với thông tin trong bài.
Câu 2: TRUE
- Dạng câu hỏi: True/False/Not Given
- Từ khóa: overwatering, root diseases
- Vị trí trong bài: Đoạn 2, dòng 4-5
- Giải thích: Bài viết nói rõ “overwatering… can also damage crops by creating waterlogged soil conditions and encouraging root diseases”. Thông tin khớp hoàn toàn với câu hỏi.
Câu 3: TRUE
- Dạng câu hỏi: True/False/Not Given
- Từ khóa: AI irrigation systems, sensors, soil moisture, temperature
- Vị trí trong bài: Đoạn 3, dòng 2-4
- Giải thích: “These systems use sensors… to continuously monitor multiple factors including soil moisture levels, temperature, humidity” – khớp chính xác với thông tin câu hỏi.
Câu 6: 50 percent / 50%
- Dạng câu hỏi: Sentence Completion
- Từ khóa: reduce water use, compared to traditional methods
- Vị trí trong bài: Đoạn 5, dòng 2-3
- Giải thích: “AI-managed irrigation systems can reduce water consumption by 30 to 50 percent compared to traditional methods” – đề hỏi mức giảm tối đa nên lấy số 50 percent.
Câu 10: B
- Dạng câu hỏi: Multiple Choice
- Từ khóa: main problem, fixed-schedule irrigation
- Vị trí trong bài: Đoạn 2, dòng 1-3
- Giải thích: Bài nói “water applied at predetermined times regardless of actual plant needs” – không xem xét nhu cầu thực tế của cây, khớp với đáp án B. Các đáp án khác không được nhắc đến trong bài.
Câu 13: C
- Dạng câu hỏi: Multiple Choice
- Từ khóa: future AI systems, being developed
- Vị trí trong bài: Đoạn cuối, dòng 2-3
- Giải thích: “Predictive models are being created that can forecast plant water needs days in advance” – khớp với đáp án C. Đáp án A, B, D không được đề cập hoặc sai với nội dung bài.
Passage 2 – Giải Thích
Câu 14: D
- Dạng câu hỏi: Matching Information
- Từ khóa: factors analyzed, predict pipe failures
- Vị trí trong bài: Section D, đoạn đầu
- Giải thích: Section D nói về “Predictive Maintenance” và liệt kê các yếu tố như “pipe age, material composition, soil conditions, pressure fluctuations, and historical weather patterns” được phân tích để dự đoán hỏng hóc.
Câu 15: A
- Dạng câu hỏi: Matching Information
- Từ khóa: statistics, water loss, global distribution systems
- Vị trí trong bài: Section A
- Giải thích: Section A có tiêu đề “The Scale of the Problem” và cung cấp số liệu thống kê như “20 to 30 percent” ở các nước phát triển và “more than 50 percent” ở một số thành phố đang phát triển.
Câu 19: flow rates
- Dạng câu hỏi: Summary Completion
- Từ khóa: sensors monitor, pressure and…
- Vị trí trong bài: Section C, đoạn đầu
- Giải thích: “These sensors monitor pressure, flow rates, water quality…” – từ cần điền là “flow rates”.
Câu 22: three meters
- Dạng câu hỏi: Summary Completion
- Từ khóa: Barcelona, locate leaks
- Vị trí trong bài: Section C, cuối đoạn 2
- Giải thích: “managed to pinpoint leaks to within three meters in a network spanning hundreds of kilometers”.
Câu 23: B
- Dạng câu hỏi: Multiple Choice
- Từ khóa: percentage water lost, developed nations
- Vị trí trong bài: Section A, dòng 2-3
- Giải thích: “In developed nations, an estimated 20 to 30 percent of treated water never reaches its destination” – đáp án B chính xác.
Câu 25: A
- Dạng câu hỏi: Multiple Choice
- Từ khóa: Taiwan semiconductor plant, pay for itself
- Vị trí trong bài: Section E, câu cuối
- Giải thích: “The system paid for itself within eight months” – đáp án A.
Passage 3 – Giải Thích
Câu 27: C
- Dạng câu hỏi: Multiple Choice
- Từ khóa: defining objectives, particularly difficult
- Vị trí trong bài: Đoạn 3, giữa đoạn
- Giải thích: Bài viết nêu rõ “These questions lack purely technical answers and instead reflect value judgments that different societies and stakeholders answer divergently” – việc định nghĩa mục tiêu khó vì liên quan đến phán đoán giá trị, không phải vấn đề kỹ thuật thuần túy.
Câu 28: B
- Dạng câu hỏi: Multiple Choice
- Từ khóa: Murray-Darling Basin, demonstrates
- Vị trí trong bài: Đoạn 4, câu cuối
- Giải thích: “AI systems excel at optimization within defined parameters but cannot independently determine what should be optimized or resolve conflicts between incommensurable values” – cho thấy sự tinh vi kỹ thuật không giải quyết được xung đột giá trị.
Câu 29: C
- Dạng câu hỏi: Multiple Choice
- Từ khóa: algorithmic authority, refers to
- Vị trí trong bài: Đoạn 5, giữa đoạn
- Giải thích: Thuật ngữ được định nghĩa là “algorithmic outputs presented as objective and scientific” và có thể “disenfranchise stakeholders who lack technical expertise to critique” – tức là cách đầu ra AI được coi là khách quan và khó thách thức.
Câu 30: B
- Dạng câu hỏi: Multiple Choice
- Từ khóa: AI optimizes from current conditions, problem
- Vị trí trong bài: Đoạn 6, giữa đoạn
- Giải thích: “An AI system optimizing from current baseline conditions may inadvertently perpetuate these inequities” – tối ưu hóa từ điều kiện hiện tại có thể duy trì bất bình đẳng.
Câu 32: satellite-based Earth observation platforms
- Dạng câu hỏi: Sentence Completion
- Từ khóa: integrate data from, meteorological sensors and
- Vị trí trong bài: Đoạn 2, dòng 2-3
- Giải thích: “These systems integrate disparate data streams from meteorological sensors, satellite-based Earth observation platforms, groundwater monitoring networks…”
Câu 37: NO
- Dạng câu hỏi: Yes/No/Not Given
- Vị trí trong bài: Các đoạn 7, 9
- Giải thích: Tác giả không đồng ý với quan điểm này. Bài viết nêu rõ nhiều hạn chế của AI như “vulnerabilities to unexpected conditions”, “performing poorly when confronted with situations substantially different from their training data”, và ví dụ về sự cố ở Pacific Northwest. Tác giả cho rằng AI có nhiều ưu điểm nhưng không phải vượt trội trong mọi tình huống.
Câu 38: YES
- Dạng câu hỏi: Yes/No/Not Given
- Vị trí trong bài: Đoạn 8
- Giải thích: Tác giả đồng ý: “the centralization of water management through AI systems creates potential vulnerabilities to catastrophic failure” và “a regional AI system… represents a single point of failure”.
Câu 40: NOT GIVEN
- Dạng câu hỏi: Yes/No/Not Given
- Vị trí trong bài: Đoạn cuối
- Giải thích: Bài viết chỉ nói “currently under consideration” (đang được xem xét), không nói đã được thực hiện (implemented). Do đó không có thông tin về việc EU đã áp dụng hay chưa.
Từ Vựng Quan Trọng Theo Passage
Passage 1 – Essential Vocabulary
| Từ vựng | Loại từ | Phiên âm | Nghĩa tiếng Việt | Ví dụ từ bài | Collocation |
|---|---|---|---|---|---|
| technological revolution | noun phrase | /ˌteknəˈlɒdʒɪkəl ˌrevəˈluːʃən/ | cuộc cách mạng công nghệ | Urban agriculture is experiencing a technological revolution | undergo/experience a technological revolution |
| artificial intelligence | noun phrase | /ˌɑːtɪˈfɪʃəl ɪnˈtelɪdʒəns/ | trí tuệ nhân tạo | artificial intelligence playing a central role | develop/deploy artificial intelligence |
| water waste | noun phrase | /ˈwɔːtə weɪst/ | lãng phí nước | addressing water waste challenges | reduce/prevent water waste |
| fixed schedules | noun phrase | /fɪkst ˈʃedjuːlz/ | lịch trình cố định | irrigation follows fixed schedules | operate on/follow fixed schedules |
| overwatering | noun/gerund | /ˌəʊvəˈwɔːtərɪŋ/ | tưới nước quá mức | overwatering wastes water resources | avoid/prevent overwatering |
| waterlogged | adjective | /ˈwɔːtəlɒɡd/ | úng nước, ngập úng | creating waterlogged soil conditions | waterlogged soil/fields |
| root diseases | noun phrase | /ruːt dɪˈziːzɪz/ | bệnh rễ cây | encouraging root diseases | prevent/control root diseases |
| stunted growth | noun phrase | /stʌntɪd ɡrəʊθ/ | sự phát triển còi cọc | leading to stunted growth | cause/result in stunted growth |
| AI-powered | adjective | /eɪ aɪ ˈpaʊəd/ | được vận hành bởi AI | AI-powered irrigation systems | AI-powered systems/technology |
| soil moisture levels | noun phrase | /sɔɪl ˈmɔɪstʃə ˈlevəlz/ | mức độ ẩm đất | monitor soil moisture levels | measure/track soil moisture levels |
| machine learning algorithms | noun phrase | /məˈʃiːn ˈlɜːnɪŋ ˈælɡərɪðəmz/ | thuật toán học máy | machine learning algorithms improve accuracy | develop/train machine learning algorithms |
| adaptive capability | noun phrase | /əˈdæptɪv ˌkeɪpəˈbɪləti/ | khả năng thích ứng | the system’s adaptive capability | enhance/demonstrate adaptive capability |
Passage 2 – Essential Vocabulary
| Từ vựng | Loại từ | Phiên âm | Nghĩa tiếng Việt | Ví dụ từ bài | Collocation |
|---|---|---|---|---|---|
| pressing challenges | noun phrase | /ˈpresɪŋ ˈtʃælɪndʒɪz/ | thách thức cấp bách | one of the most pressing challenges | face/address pressing challenges |
| water-stressed regions | noun phrase | /ˈwɔːtə strest ˈriːdʒənz/ | khu vực thiếu nước | living in water-stressed regions | water-stressed areas/countries |
| distribution systems | noun phrase | /ˌdɪstrɪˈbjuːʃən ˈsɪstəmz/ | hệ thống phân phối | water waste in distribution systems | maintain/upgrade distribution systems |
| game-changing technology | noun phrase | /ɡeɪm ˈtʃeɪndʒɪŋ tekˈnɒlədʒi/ | công nghệ thay đổi cuộc chơi | emerging as a game-changing technology | develop/introduce game-changing technology |
| manual inspection | noun phrase | /ˈmænjuəl ɪnˈspekʃən/ | kiểm tra thủ công | relied on manual inspection | conduct/perform manual inspection |
| woefully inadequate | adjective phrase | /ˈwəʊfəli ɪnˈædɪkwət/ | không đủ một cách đáng buồn | woefully inadequate for detecting | prove/remain woefully inadequate |
| acoustic signatures | noun phrase | /əˈkuːstɪk ˈsɪɡnətʃəz/ | dấu hiệu âm thanh | monitor acoustic signatures | detect/analyze acoustic signatures |
| data streams | noun phrase | /ˈdeɪtə striːmz/ | luồng dữ liệu | analyze data streams continuously | process/integrate data streams |
| anomalous patterns | noun phrase | /əˈnɒmələs ˈpætənz/ | các mẫu bất thường | searching for anomalous patterns | identify/detect anomalous patterns |
| predictive maintenance | noun phrase | /prɪˈdɪktɪv ˈmeɪntənəns/ | bảo trì dự đoán | AI’s capacity for predictive maintenance | implement/schedule predictive maintenance |
| risk assessments | noun phrase | /rɪsk əˈsesmənts/ | đánh giá rủi ro | generate risk assessments | conduct/perform risk assessments |
| proactive repairs | noun phrase | /prəʊˈæktɪv rɪˈpeəz/ | sửa chữa chủ động | prioritize proactive repairs | carry out/schedule proactive repairs |
| legacy infrastructure | noun phrase | /ˈleɡəsi ˈɪnfrəstrʌktʃə/ | cơ sở hạ tầng kế thừa | operate with legacy infrastructure | upgrade/replace legacy infrastructure |
| interoperability issues | noun phrase | /ˌɪntərˌɒpərəˈbɪləti ˈɪʃuːz/ | vấn đề tương tác | interoperability issues when integrating | address/resolve interoperability issues |
| cybersecurity concerns | noun phrase | /ˈsaɪbəsɪˌkjʊərəti kənˈsɜːnz/ | mối quan ngại an ninh mạng | cybersecurity concerns represent consideration | raise/address cybersecurity concerns |
Passage 3 – Essential Vocabulary
| Từ vựng | Loại từ | Phiên âm | Nghĩa tiếng Việt | Ví dụ từ bài | Collocation |
|---|---|---|---|---|---|
| paradigm shift | noun phrase | /ˈpærədaɪm ʃɪft/ | sự thay đổi mô hình | represents a paradigm shift | undergo/experience a paradigm shift |
| intersection | noun | /ˌɪntəˈsekʃən/ | giao điểm | at the intersection of considerations | at the intersection of |
| governance | noun | /ˈɡʌvənəns/ | quản trị | questions about governance | improve/enhance governance |
| algorithmic decision-making | noun phrase | /ˌælɡəˈrɪðmɪk dɪˈsɪʒən ˈmeɪkɪŋ/ | ra quyết định bằng thuật toán | nature of algorithmic decision-making | algorithmic decision-making processes |
| disparate data streams | noun phrase | /ˈdɪspərət ˈdeɪtə striːmz/ | các luồng dữ liệu khác nhau | integrate disparate data streams | combine/integrate disparate data streams |
| deep learning neural networks | noun phrase | /diːp ˈlɜːnɪŋ ˈnjʊərəl ˈnetwɜːks/ | mạng nơ-ron học sâu | through deep learning neural networks | train/develop deep learning neural networks |
| multidimensional models | noun phrase | /ˌmʌltɪdaɪˈmenʃənəl ˈmɒdəlz/ | các mô hình đa chiều | constructs multidimensional models | build/create multidimensional models |
| multivariate optimization | noun phrase | /ˌmʌltiˈveəriət ˌɒptɪmaɪˈzeɪʃən/ | tối ưu hóa đa biến | a multivariate optimization problem | solve/address multivariate optimization |
| objective functions | noun phrase | /əbˈdʒektɪv ˈfʌŋkʃənz/ | các hàm mục tiêu | require explicit objective functions | define/specify objective functions |
| intractable challenges | noun phrase | /ɪnˈtræktəbəl ˈtʃælɪndʒɪz/ | thách thức khó giải quyết | presents intractable challenges | face/encounter intractable challenges |
| value judgments | noun phrase | /ˈvæljuː ˈdʒʌdʒmənts/ | phán đoán giá trị | reflect value judgments | make/involve value judgments |
| econometric modeling | noun phrase | /ɪˌkɒnəˈmetrɪk ˈmɒdəlɪŋ/ | mô hình kinh tế lượng | derived from econometric modeling | use/apply econometric modeling |
| stakeholder consultation | noun phrase | /ˈsteɪkhəʊldə ˌkɒnsəlˈteɪʃən/ | tham vấn các bên liên quan | through stakeholder consultation processes | conduct/facilitate stakeholder consultation |
| technocratic framework | noun phrase | /ˌteknəˈkrætɪk ˈfreɪmwɜːk/ | khuôn khổ kỹ trị | creating a technocratic framework | establish/operate within a technocratic framework |
| algorithmic authority | noun phrase | /ˌælɡəˈrɪðmɪk ɔːˈθɒrəti/ | quyền lực thuật toán | phenomenon termed algorithmic authority | challenge/question algorithmic authority |
| legacy effects | noun phrase | /ˈleɡəsi ɪˈfekts/ | tác động kế thừa | reflect legacy effects of colonialism | address/overcome legacy effects |
| incremental improvement | noun phrase | /ˌɪŋkrɪˈmentəl ɪmˈpruːvmənt/ | cải tiến từng bước | tends toward incremental improvement | achieve/focus on incremental improvement |
| catastrophic failure | noun phrase | /ˌkætəˈstrɒfɪk ˈfeɪljə/ | hỏng hóc thảm khốc | vulnerabilities to catastrophic failure | prevent/avoid catastrophic failure |
| hybrid governance models | noun phrase | /ˈhaɪbrɪd ˈɡʌvənəns ˈmɒdəlz/ | mô hình quản trị lai | involve hybrid governance models | develop/implement hybrid governance models |
| algorithmic transparency | noun phrase | /ˌælɡəˈrɪðmɪk trænsˈpærənsi/ | tính minh bạch thuật toán | mandatory algorithmic transparency requirements | ensure/promote algorithmic transparency |
Học viên luyện thi IELTS Reading với chủ đề AI quản lý nước và giảm lãng phí tài nguyên
Kết Luận
Chủ đề “AI for reducing water waste” không chỉ mang tính thời sự cao mà còn phản ánh xu hướng ra đề của IELTS Reading trong những năm gần đây – kết hợp giữa công nghệ, môi trường và các vấn đề toàn cầu. Qua bộ đề thi mẫu này, bạn đã được trải nghiệm đầy đủ ba mức độ khó từ Easy đến Hard, giúp bạn làm quen với cách thông tin được trình bày và paraphrase trong đề thi thật.
Ba passages trong đề thi đã cung cấp góc nhìn toàn diện về ứng dụng AI trong quản lý nước: từ nông nghiệp đô thị (Passage 1), đến các hệ thống công nghiệp quy mô lớn (Passage 2), và cuối cùng là những phân tích sâu sắc về mặt đạo đức và xã hội (Passage 3). Sự đa dạng này giúp bạn rèn luyện khả năng xử lý nhiều loại văn bản khác nhau trong cùng một bài thi.
Đáp án chi tiết kèm giải thích đã chỉ ra cách xác định từ khóa, tìm thông tin trong bài và áp dụng kỹ thuật paraphrase – những kỹ năng cốt lõi để đạt band điểm cao. Bảng từ vựng theo từng passage cung cấp nguồn học liệu quý giá, giúp bạn mở rộng vốn từ vựng học thuật cần thiết cho kỳ thi.
Hãy làm lại đề thi này nhiều lần, phân tích kỹ những câu bạn làm sai, và chú ý đến cách thông tin được paraphrase giữa câu hỏi và passage. Đây chính là con đường hiệu quả nhất để cải thiện kỹ năng IELTS Reading của bạn. Chúc bạn ôn tập tốt và đạt được band điểm mục tiêu!
