📚 Identifying errors in algorithms | 识别算法中的错误
Algorithms are step‑by‑step procedures or formulas for solving a problem. In IGCSE Science, you might write simple algorithms to process experimental data, simulate population growth, or calculate chemical yields. Even a small mistake in an algorithm can lead to completely wrong conclusions. Learning to identify and fix these errors is an essential skill, not only in computing but throughout the scientific method.
算法是解决问题的一步一步的过程或公式。在IGCSE科学中，你可能会编写简单的算法来处理实验数据、模拟种群增长或计算化学产率。算法中哪怕一个微小的错误，都可能导致完全错误的结论。学会识别并修正这些错误不仅对计算机科学至关重要，对整套科学方法而言也是核心能力。
1. What is an algorithm? | 什么是算法？
An algorithm is a precise sequence of instructions designed to perform a specific task. In science, these tasks could be sorting measured values, averaging results, or deciding whether a reaction has reached equilibrium. Algorithms can be expressed in plain English, pseudocode, flowcharts, or actual programming languages.
算法是旨在执行某个特定任务的精确指令序列。在科学中，这些任务可能是排序测量值、计算结果的平均值，或判断一个反应是否达到平衡。算法可以用简单的英语、伪代码、流程图或者真实的编程语言来表达。
For example, a simple algorithm to find the mean of three repeated readings from a calorimetry experiment looks like this:
例如，一个用于求热量计实验的三次重复读数平均值的简单算法如下：
- Input temp1, temp2, temp3
- sum ← temp1 + temp2 + temp3
- mean ← sum ÷ 3
- Output mean
2. Algorithms in everyday science | 日常科学中的算法
Algorithms appear whenever you follow a written method in a lab or use a spreadsheet formula. A titration calculation algorithm might read burette readings, apply a correction factor, and compute concentration. A biology fieldwork algorithm could estimate a population using the Lincoln index. In physics, an algorithm might simulate motion step by step.
每当你在实验室里按照书面方法操作，或使用电子表格公式时，算法都会出现。一个滴定计算算法可能会读取滴定管读数、应用修正因子，并计算出浓度。一个生物学实地考察算法可能利用林肯指数估算种群数量。在物理中，算法可能一步一步地模拟运动过程。
Because scientific results depend on accurate algorithms, identifying any flaw in the logic or data handling is as critical as calibrating a measuring instrument.
正因为科学结果依赖于准确的算法，识别逻辑或数据处理中的任何缺陷，就像校准测量仪器一样至关重要。
3. Types of algorithmic errors | 算法错误的类型
Errors in algorithms generally fall into three main categories: syntax errors, logic errors, and runtime errors. Syntax errors break the rules of the programming language or expression; logic errors produce an incorrect output even when the algorithm runs smoothly; runtime errors cause the algorithm to fail during execution, often because of illegal operations.
算法中的错误通常分为三大类：语法错误、逻辑错误和运行时错误。语法错误违反了编程语言或表达式的规则；逻辑错误即使算法顺利运行，也会产生不正确的输出；运行时错误会在执行过程中导致算法崩溃，常常是因为非法操作。
Understanding these categories helps you systematically trace and correct faults, which is a vital part of scientific computing and data analysis.
理解这些分类有助于你系统地追踪和改正故障，这对于科学计算和数据分析都是极其重要的能力。
4. Syntax errors – breaking the rules | 语法错误——违反规则
A syntax error occurs when you write an instruction that the computer or formula parser does not understand. Examples include misspelling a function name, missing a parenthesis, or using an invalid symbol. In a chemistry spreadsheet, typing =AVERGE(B2:B10) instead of =AVERAGE(B2:B10) is a syntax error that prevents the formula from working.
当你写出一条计算机或公式解析器无法理解的指令时，就产生了语法错误。例如，拼错函数名、缺少括号或使用了无效符号。在化学电子表格中，输入=AVERGE(B2:B10)而不是=AVERAGE(B2:B10)，就是一个阻止公式正常工作的语法错误。
Modern development environments highlight syntax errors immediately, but when writing pseudocode or flowcharting by hand, you must carefully check spelling, punctuation, and reserved words.
现代开发环境会立刻高亮语法错误，但在手写伪代码或绘制流程图时，你必须仔细检查拼写、标点符号和保留字。
5. Logic errors – hidden traps | 逻辑错误——隐蔽的陷阱
Logic errors are the most dangerous because the algorithm runs without any error message, yet the result is wrong. A typical logic error is using the wrong mathematical operator. Suppose you need the average acceleration and write acceleration = (v – u) ÷ t but mistakenly code acceleration = v – u ÷ t, which performs division before subtraction, giving an erroneous value.
逻辑错误最为危险，因为算法能够无误运行，但结果却是错误的。一个典型的逻辑错误是使用了错误的数学运算符。假设你需要计算平均加速度，公式为加速度 = (v – u) ÷ t，但不小心编写成加速度 = v – u ÷ t，这会先执行除法再执行减法，从而得出错误的值。
Another common logic error is an infinite loop that never terminates, or a conditional statement with an incorrect comparison operator such as writing > instead of ≥ when checking if a temperature exceeds a threshold.
另一个常见的逻辑错误是永不终止的无限循环，或者条件语句中使用了错误的比较运算符，例如在检查温度是否超过阈值时把>写成了≥。
To detect logic errors, you must test the algorithm with known inputs and check whether the outputs match expected values.
要检测逻辑错误，你必须用已知的输入测试算法，并检查输出是否与预期值吻合。
6. Runtime errors – unexpected stop | 运行时错误——意外中止
Runtime errors appear while the algorithm is executing. Examples include dividing by zero, attempting to take the square root of a negative number in a real‑valued model, or accessing an element outside the bounds of an array. In a population growth simulation, if the birth rate suddenly becomes zero, a division in the growth formula could crash the program.
运行时错误在算法执行过程中出现。例如，除以零、在实数模型中试图对负数开平方，或者访问数组边界之外的元素。在一个种群增长模拟中，如果出生率突然变为零，增长公式中的除法可能会导致程序崩溃。
Defensive design can prevent many runtime errors: always validate inputs, check for zero divisors before division, and use conditions to keep variables within realistic scientific ranges.
防御性设计可以防止许多运行时错误：始终验证输入，在做除法前检查除数是否为零，并使用条件将变量保持在现实的科学范围内。
7. Identifying errors through systematic testing | 通过系统测试识别错误
Testing is the most reliable way to uncover hidden faults. For IGCSE Science algorithms, you should create a test plan that includes normal data, boundary data, and invalid data. Normal data are typical values from a laboratory investigation. Boundary data test the extremes, for instance maximum and minimum possible temperatures. Invalid data, like a negative mass, should be handled gracefully by the algorithm.
测试是发现隐藏错误的最可靠方法。对于IGCSE科学算法，你应该制定一个包含正常数据、边界数据和无效数据的测试计划。正常数据是实验室探究中的典型值。边界数据测试极端情况，例如最高和最低可能的温度。无效数据，比如负的质量，应由算法体面地处理。
Every test should state the expected outcome before running the algorithm, so you can immediately spot a discrepancy.
每次测试都应该在运行算法之前说明预期结果，这样就能立即发现差异。
| Test type 测试类型 | Input example 输入示例 | Expected output 预期输出 |
|---|---|---|
| Normal | Temp = 25 °C, 27 °C, 26 °C | Mean = 26 °C |
| Boundary | Temp = 0 °C, 100 °C | Still valid (specific range) |
| Invalid | Temp = -5 °C (impossible in context) | Error message or rejection |
8. Tracing algorithms with trace tables | 使用跟踪表追踪算法
A trace table is a manual debugging tool where you write down the state of every variable at each step of the algorithm. This technique reveals exactly where a variable takes an unexpected value. For a momentum calculation p = m × v, a trace table would list m, v, and p after each line, making it easy to see if the multiplication was carried out incorrectly.
跟踪表是一种手动调试工具，你在算法的每一步记录下每个变量的状态。这一技术可以精确揭示变量在何处出现了不符合预期的值。对于动量计算p = m × v，一个跟踪表会在每一行后列出m、v和p，从而很容易看出乘法是否执行错误。
Tracing is especially useful when checking loops. If an algorithm repeats too many or too few times, the trace table will show the loop counter exceeding the intended limit.
追踪在检查循环时尤其有用。如果算法重复次数过多或过少，跟踪表会显示循环计数器超出了预期的界限。
The skill of tracing is assessed in IGCSE questions, so practice building trace tables for simple algorithms like finding the highest temperature from a set of readings.
跟踪技能在IGCSE考试中会有所考查，所以要常练习为简单算法构建跟踪表，例如从一组读数中找出最高温度。
9. Checking boundary conditions and edge cases | 检查边界条件和边缘情况
Boundary conditions are values at which the algorithm’s behaviour changes, such as a threshold temperature triggering a chemical process or a pH value separating acids from bases. Edge cases are extreme but valid situations, like a very small population or zero concentration. Algorithms often fail at these points because the programmer only considered typical values.
边界条件是算法行为发生改变的数值，例如触发某个化学过程的阈值温度，或者区分酸和碱的pH值。边缘情况是极端但有效的情形，比如极小的种群或零浓度。算法经常在这些点失败，因为编程者只考虑了典型数值。
In an enzyme activity algorithm, you might calculate rate = change in product ÷ time. If time = 0 at the very start, this is an edge case that must be avoided by starting measurement after a delay. Always pad your algorithm with checks for such special cases.
在酶活性算法中，你可能会计算速率 = 产物变化量 ÷ 时间。如果一开始时间 = 0，这就是一个边缘情况，必须通过在延迟后再开始测量来避免。始终为这类特殊情况添加检查来保护你的算法。
10. Interpreting error messages | 解读错误信息
When an algorithm fails with an error code or message, read it carefully: it often tells you the line number and the type of mistake. In a spreadsheet, the error #DIV/0! immediately points to a division by zero. In a biology simulation written in Python, IndexError reveals that the program tried to read beyond the available data list. Learning to decode these messages speeds up debugging enormously.
当算法失败并给出错误代码或信息时，仔细阅读：它通常会告诉你行号和错误类型。在电子表格中，错误#DIV/0!立即指向除以零。在用Python编写的生物学模拟中，IndexError揭示程序试图读取超出可用数据列表的项。学会解读这些信息能极大加快调试速度。
Never ignore an error message even if the algorithm seems to continue. Hidden errors can corrupt your scientific dataset and lead to false conclusions about an experiment.
即使算法看似仍然运行，也不要忽视任何错误信息。隐藏的错误会破坏你的科学数据集，并导致对实验的错误结论。
11. Peer review and code walkthrough | 同行评审与代码走查
Just as you peer-review a lab report, you can ask a classmate to inspect your algorithm. A fresh pair of eyes often spots a misplaced operator or a missing step that you have overlooked. A structured walkthrough involves explaining each line out loud while pointing to the relevant variable values. This oral method forces you to confront your own assumptions.
就像你会同行评审一份实验报告一样，你可以请同学检查你的算法。一双陌生的眼睛常常能发现你忽略的错位运算符或遗漏的步骤。结构化的走查包括一面指着相应的变量值一面大声解释每一行代码。这种口头方法迫使你正视自己的假设。
In IGCSE group projects, building algorithms together and checking each other’s logic improves the overall quality of the scientific investigation and reduces the risk of random errors creeping into the data analysis.
在IGCSE团队项目中，一起搭建算法并相互检查逻辑，既能提高科学探究的整体质量，也能降低随机差错潜入数据分析的风险。
12. Correcting errors and re‑verifying | 纠正错误并重新验证
Once you locate and fix an error, you must re‑run all previous tests to make sure the correction did not introduce new faults. This is called regression testing. After adjusting a formula in a physics simulation, for example, confirm that the outputs for a set of standard benchmarks still agree with known physical laws.
一旦定位并修正了一个错误，你必须重新运行所有之前的测试，以确保修正没有引入新的错误。这被称为回归测试。例如，在调整了物理模拟中的某个公式后，要确认一组标准基准的输出仍然符合已知的物理定律。
Document the error and its solution in your scientific logbook; this helps you avoid repeating the same mistake and shows examiners that you understand the debugging process as part of scientific methodology.
在你的科学日志中记录错误及其解决方案；这有助于避免重蹈覆辙，并向考官展示你理解作为科学方法一部分的调试过程。
Remember: an algorithm is correct only when it produces accurate results for all foreseeable inputs, handles errors safely, and is clearly understood by other scientists.
请记住：只有当算法对所有可预见的输入都能产生准确结果、安全地处理错误、并能被其他科学家明确理解时，它才是正确的。
Published by TutorHao | Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导