Computer error analysis method and equipment based on fuzzy logic

Abstract

The invention relates to the technical field of computer science and data processing technologies, in particular to a computer error analysis method and device based on fuzzy logic, and the method comprises the steps: S1, obtaining basic data: obtaining current original data of a computer; s2, preprocessing the data; performing cleaning and integration processing on the collected original data, removing noise and abnormal values, and mapping the data to an interval [0, 1] for subsequent fuzzification processing; s3, performing fuzzy processing on the data in the S2; a fuzzy set and a membership function are defined, a normalized data set is formed, three groups of fuzzy error sets are formed through fuzzy definition, a fuzzy rule is established, and finally an error condition and an analysis report are obtained, definition and operation of the fuzzy set are relatively simple, and complex mathematical derivation and calculation do not need to be carried out. The reasoning process is more intuitive and efficient due to the application of the fuzzy rule, and the calculation amount is greatly reduced by converting a complex problem into matching and reasoning of the fuzzy rule.

Description

Technical Field

[0001] The present invention relates to the technical field of computer science and data processing technology, and in particular to a computer error analysis method and equipment based on fuzzy logic. Background Art

[0002] In today's digital age, computer technology is widely used in various fields, from scientific research, industrial production to daily life, data processing and analysis have become key links. Error analysis, as an important means to ensure data accuracy and system reliability, has always been a research hotspot in the computer field.

[0003] Traditional error analysis methods are mainly based on precise mathematical models and statistical theories. These methods perform well when processing data with clear rules and certainty. For example, in the processing of classical physical experimental data, by establishing precise mathematical equations and applying statistical analysis methods, the measurement error can be accurately calculated and the experimental results can be evaluated. In the quality control link of industrial production, methods based on statistical process control (SPC) can timely detect abnormal fluctuations in the production process based on the statistical characteristics of production data, thereby ensuring the stability of product quality.

[0004] However, with the continuous development of science and technology, the data processed by computers has become more and more complex and diverse, and the limitations of traditional error analysis methods have gradually emerged.

[0005] From the perspective of data sources and scale, today's data sources are extensive, including various sensors, network devices, social media, etc. These data are not only diverse, but also large in scale. Traditional error analysis methods have high computational complexity and low efficiency when processing large-scale data. At the same time, the data formats and quality of different data sources are also different, which further increases the difficulty of error analysis.

[0006] In actual application scenarios, many systems have high requirements for real-time and adaptability. For example, in self-driving cars, it is necessary to perform error analysis on the data collected by sensors in real time to ensure the safe driving of the vehicle. Traditional methods are difficult to meet real-time requirements because they need to establish accurate models and perform complex calculations. Moreover, when the operating environment of the system changes, traditional methods need to readjust model parameters, which lacks flexibility and adaptability.

[0007] In addition, in some complex systems, the relationships between various factors are often nonlinear and complex. For example, the various biological and environmental factors in the ecosystem interact with each other to form a complex ecological network. Traditional error analysis methods are difficult to accurately describe these complex relationships, resulting in the neglect of some important factors when analyzing errors, thus affecting the reliability of the analysis results. Summary of the invention

[0008] Technical issues solved In view of the above shortcomings of the prior art, the present invention provides a computer error analysis method and device based on fuzzy logic, which can obtain error analysis results in a shorter time when processing large-scale data, while reducing the demand for hardware resources. This not only saves computing time, but also reduces the operating cost of the system, so that error analysis can be implemented in a wider range of devices and environments, and can effectively solve the problem that the prior art accurately describes these complex relationships, resulting in the neglect of some important factors when analyzing errors, thereby affecting the reliability of the analysis results.

[0009] Technical Solution To achieve the above objectives, the present invention is implemented through the following technical solutions: The present invention provides a computer error analysis method based on fuzzy logic, including a computer error analysis method based on fuzzy logic, comprising: S1: Get basic data: Get the current computer raw data; S2: Preprocess the data; clean and integrate the collected raw data, remove noise and outliers, and map the data to the [0, 1] interval for subsequent fuzzy processing; S3: fuzzify the data in S2; define fuzzy sets and membership functions, and convert the preprocessed data into fuzzy variables; S4: Establish a fuzzy data rule base; and establish a series of fuzzy rules; S5: Fuzzy reasoning: Based on the input fuzzy variables and fuzzy rule base, the fuzzy reasoning algorithm is used to perform reasoning to obtain the fuzzy set of output variables; S6: Defuzzification: convert the output fuzzy set obtained by fuzzy reasoning into an accurate value; S7: Output of error analysis results. According to the defuzzified results, the error is analyzed and evaluated, and an error analysis report is output to provide a basis for subsequent decision-making.

[0010] Furthermore, in S2, data cleaning includes processing missing values, processing outliers and processing duplicate values. When processing missing values: you can directly delete samples or features containing missing values, or you can use mean filling, that is, for numerical data, calculate the mean in the database to fill missing values; when processing outliers: based on statistical methods, calculate the degree of deviation of each data point, with the standard deviation as the value, and usually regard data points greater than the threshold as outliers, and find out the outliers through calculation. When processing duplicate values: for completely repeated samples, you can directly delete the duplicates and only keep one unique sample, and integrate the cleaned data samples, normalize the data sources with the same meaning but different expressions, and obtain the processed data set, and realize data conversion through the Min-Max standardization method, and scale the processed data set to the [0, 1] interval, which is specifically expressed as: , where Xmin and Xmax are the minimum and maximum values ​​of the data respectively, and the length of the data vector is normalized to 1 to form a normalized data set.

[0011] Furthermore, when dealing with outliers, the z-score method can be used for calculation, which is expressed as: , where X is the original data point, U is the mean of the data set, and Σ is the standard deviation of the data set.

[0012] Furthermore, the normalized data set in S2 is fuzzily defined and divided into three fuzzy sets: "small error", "medium error" and "large error". The membership function of the error is: , where e represents data error, and based on experience, a fuzzy data rule base is established to establish fuzzy rules.

[0013] Furthermore, in S4, the fuzzy rule is obtained as follows: If: If the data error is small and the error change rate is small, the error state is stable.

[0014] If: If the data error is medium error and the error change rate is medium, then the error status is attention-needed.

[0015] If: If the data error is a large error and the error change rate is large, the error state is serious; according to the fuzzy rule base and Mamdani reasoning algorithm, the fuzzy set of the output error state is obtained. Furthermore, the reasoning algorithm process in S4 is as follows: in the normalized data, "small error", "medium error" and "high error" are set as fuzzy sets, and their membership in each fuzzy set is determined according to the specific value of the input, and then the input fuzzy set is matched with the rules in the pre-established fuzzy rule base, and the matching degree between the input fuzzy set and the rule before is compared to determine the activated rules. For the matched rules, the activation strength of the rules is calculated according to the matching degree between the input fuzzy set and the rule before, and the output fuzzy sets of all activated rules are synthesized. The synthesis uses the maximum method of taking the union of the output fuzzy sets of all activated rules as the final output to obtain the final output fuzzy set, forming the fuzzy data set in S5.

[0016] Further, in S6, the fuzzy data set in S5 is calculated using the centroid method. , where μ(x) is its membership function, [a, b] is its domain, and X* is its exact value, and then the precise error state value is obtained.

[0017] Furthermore, in S7, the error condition of the current data is determined based on the error state value after defuzzification in S6, and an error analysis report is output.

[0018] To achieve the above objectives, the present invention also provides a computer error analysis device based on fuzzy logic to achieve: Includes: Data acquisition module: used to obtain basic data; Analysis module: cleans and integrates data; Data fuzzification processing module: convert data into fuzzy variables. Data reasoning module: use fuzzy algorithm to reason about fuzzy edge beams and obtain fuzzy sets. Data conversion module, converts fuzzy sets into precise values, The result output module analyzes and evaluates the errors and outputs error analysis reports.

[0019] Beneficial Effects Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects: The present invention pre-processes the data, maps the data, and then performs fuzzification processing. The use of fuzzy logic can effectively process the uncertainty and fuzziness of the data, avoids the problem of inaccurate error analysis results caused by inaccurate data in traditional methods, and improves the accuracy of error analysis. In the present invention, for the data cleaning part, classification processing is performed, and the only sample is retained. A normalized data set is formed through the Min-Max standardization method, and then three groups of fuzzy error sets are formed through fuzzy definition, and fuzzy rules are established. Finally, the error situation and analysis report are obtained. The definition and operation of the fuzzy set are relatively simple, and no complex mathematical derivation and calculation are required. The application of fuzzy rules also makes the reasoning process more intuitive and efficient. By converting complex problems into matching and reasoning of fuzzy rules, the amount of calculation is greatly reduced.

[0020] In the present invention, when processing large-scale data, the error analysis results can be obtained in a shorter time, while reducing the demand for hardware resources. This not only saves computing time, but also reduces the operating cost of the system, so that the error analysis can be implemented in a wider range of devices and environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a block diagram of the device structure of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] The present invention will be further described below in conjunction with the embodiments.

[0025] Example: Refer to the attached Figure 1 , computer error analysis method based on fuzzy logic, including: S1: Get basic data: Get the current computer raw data; S2: Preprocess the data; clean and integrate the collected raw data, remove noise and outliers, and map the data to the [0, 1] interval for subsequent fuzzy processing; S3: fuzzify the data in S2; define fuzzy sets and membership functions, and convert the preprocessed data into fuzzy variables; S4: Establish a fuzzy data rule base; and establish a series of fuzzy rules; S5: Fuzzy reasoning: Based on the input fuzzy variables and fuzzy rule base, the fuzzy reasoning algorithm is used to perform reasoning to obtain the fuzzy set of output variables; S6: Defuzzification: convert the output fuzzy set obtained by fuzzy reasoning into an accurate value; S7: Output of error analysis results: According to the defuzzified results, the error is analyzed and evaluated, and an error analysis report is output to provide a basis for subsequent decision-making; In S2, data cleaning includes processing missing values, processing outliers, and processing duplicate values. When processing missing values: you can directly delete samples or features containing missing values, or you can use mean filling, that is, for numerical data, calculate the mean in the database to fill missing values; when processing outliers: based on statistical methods, calculate the degree of deviation of each data point, with the standard deviation as the value, and usually regard data points greater than the threshold as outliers, and find out the outliers through calculation. When processing duplicate values: for completely repeated samples, you can directly delete the duplicates and only keep one unique sample. The cleaned data samples are integrated and processed, and the data sources with the same meaning but different expressions are normalized to obtain the processed data set. The data conversion is realized through the Min-Max standardization method, and the processed data set is scaled to the [0, 1] interval, which is specifically expressed as: , where Xmin and Xmax are the minimum and maximum values ​​of the data respectively, and the length of the data vector is normalized to 1 to form a normalized data set.

[0026] Preprocess the data, which is a crucial link in the fields of data mining and machine learning. It can improve the data quality, thereby improving the effect of subsequent analysis and modeling; after mapping the data, fuzzification is performed, and the use of fuzzy logic can effectively handle the uncertainty and fuzziness of the data, avoiding the problem of inaccurate error analysis results caused by inaccurate data in traditional methods, and improving the accuracy of error analysis. In the present invention, for the data cleaning part, classification processing is performed, and the only sample is retained. Through the Min-Max standardization method, a normalized data set is formed, and then through fuzzy definition, three groups of fuzzy error sets are formed, and fuzzy rules are established. Finally, the error situation and analysis report are obtained. The definition and operation of fuzzy sets are relatively simple, and no complex mathematical derivation and calculation are required. The application of fuzzy rules also makes the reasoning process more intuitive and efficient, and the amount of calculation is greatly reduced by converting complex problems into matching and reasoning of fuzzy rules.

[0027] Embodiment 2: When processing outliers, the z-score method can be used for calculation, which is expressed as: , where X is the original data point, U is the mean of the data set, and Σ is the standard deviation of the data set; The normalized data set in S2 is fuzzily defined and divided into three fuzzy sets: "small error", "medium error" and "large error". The membership function of the error is: , where e represents data error, and based on experience, a fuzzy data rule base is established to establish fuzzy rules; in S4, the fuzzy rules are obtained as follows: If: If the data error is small and the error change rate is small, the error state is stable.

[0028] If: If the data error is medium error and the error change rate is medium, then the error status is attention-needed.

[0029] If: If the data error is a large error and the error change rate is large, the error state is serious; reasoning is performed based on the fuzzy rule base and the Mamdani inference algorithm to obtain the fuzzy set of the output error state.

[0030] Embodiment 3: The reasoning algorithm process in S4 is as follows: in the normalized data, "small error", "medium error" and "high error" are set as fuzzy sets, and the membership of the specific temperature value in each fuzzy set is determined according to the input, and then the input fuzzy set is matched with the rules in the pre-established fuzzy rule base, and the matching degree between the input fuzzy set and the rule is compared to determine the activated rule. For the matched rules, the activation strength of the rule is calculated according to the matching degree between the input fuzzy set and the rule, and the output fuzzy sets of all activated rules are synthesized. The synthesis uses the maximum method of taking the union of the output fuzzy sets of all activated rules as the final output to obtain the final output fuzzy set, forming the fuzzy data set in S5. In S6, the fuzzy data set in S5 is calculated using the centroid method. , where μ(x) is its membership function, [a, b] is its domain, and X* is its exact value. Then, the precise error state value is obtained. In S7, the error of the current temperature data is judged according to the error state value after defuzzification in S6, and the error analysis report is output. When processing large-scale data, the error analysis results can be obtained in a shorter time, while reducing the demand for hardware resources. This not only saves computing time, but also reduces the operating cost of the system, so that error analysis can be implemented in a wider range of devices and environments.

[0031] Embodiment 4: Refer to the attached Figure 2 , based on the first to third embodiments, a computer error analysis device based on fuzzy logic is also introduced, which includes: Data acquisition module: used to obtain basic data; Analysis module: cleans and integrates data; Data fuzzification processing module: convert data into fuzzy variables. Data reasoning module: use fuzzy algorithm to reason about fuzzy edge beams and obtain fuzzy sets. Data conversion module, converts fuzzy sets into precise values, The result output module analyzes and evaluates the errors and outputs error analysis reports.

[0032] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A computer error analysis method based on fuzzy logic, characterized in that: include: S1: Get basic data: Get the current computer raw data; S2: preprocess the data; Clean and integrate the collected raw data, remove noise and outliers, and map the data to the [0, 1] interval for subsequent fuzzy processing; S3: Fuzzify the data in S2; define fuzzy sets and membership functions, and convert the preprocessed data into fuzzy variables; S4: Establish a fuzzy data rule base; and establish a series of fuzzy rules; S5: Fuzzy reasoning: Based on the input fuzzy variables and fuzzy rule base, the fuzzy reasoning algorithm is used to perform reasoning to obtain the fuzzy set of output variables; S6: Defuzzification: convert the output fuzzy set obtained by fuzzy reasoning into an accurate numerical value; S7: Output of error analysis results. According to the defuzzified results, the error is analyzed and evaluated, and an error analysis report is output to provide a basis for subsequent decision-making.

2. The computer error analysis method based on fuzzy logic according to claim 1, characterized in that: In S2, data cleaning includes processing missing values, processing outliers, and processing duplicate values. When processing missing values: you can directly delete samples or features containing missing values, or you can use mean filling, that is, for numerical data, calculate the mean in the database to fill missing values; when processing outliers: based on statistical methods, calculate the degree of deviation of each data point, with the standard deviation as the value, and usually regard data points greater than the threshold as outliers, and find out the outliers through calculation. When processing duplicate values: for completely repeated samples, you can directly delete the duplicates and only keep one unique sample. The cleaned data samples are integrated and processed, and the data sources with the same meaning but different expressions are normalized to obtain the processed data set. The data conversion is realized through the Min-Max normalization method, and the processed data set is scaled to the [0, 1] interval, and the data vector length is normalized to 1 to form a normalized data set.

3. The computer error analysis method based on fuzzy logic according to claim 2 is characterized in that: When dealing with outliers, the z-score method can be used for calculation.

4. The computer error analysis method based on fuzzy logic according to claim 1, characterized in that: The normalized data set in S2 is fuzzily defined and divided into three fuzzy sets: "small error", "medium error" and "large error". Based on experience, a fuzzy data rule base and fuzzy rules are established.

5. The computer error analysis method based on fuzzy logic according to claim 4 is characterized in that: In S4, the fuzzy rule is: If: If the data error is small and the error change rate is small, the error state is stable; If: If the data error is medium and the error change rate is medium, then the error status is attention-needed; If: If the data error is a large error and the error change rate is large, the error state is serious; reasoning is performed based on the fuzzy rule base and the Mamdani inference algorithm to obtain the fuzzy set of the output error state.

6. The computer error analysis method based on fuzzy logic according to claim 5, characterized in that: The reasoning algorithm process in S4 is as follows: in the normalized data, "small error", "medium error" and "high error" are set as fuzzy sets, and their membership in each fuzzy set is determined according to the specific value of the input. Then, the input fuzzy set is matched with the rules in the pre-established fuzzy rule base, and the matching degree between the input fuzzy set and the rule before is compared to determine the activated rules. For the matched rules, the activation strength of the rules is calculated according to the matching degree between the input fuzzy set and the rule before, and the output fuzzy sets of all activated rules are synthesized. The synthesis uses the maximum method of taking the union of the output fuzzy sets of all activated rules as the final output to obtain the final output fuzzy set, forming the fuzzy data set in S5.

7. The computer error analysis method based on fuzzy logic according to claim 1, characterized in that: In S6, the fuzzy data set in S5 is calculated using the centroid method, and then the accurate error state value is obtained.

8. The computer error analysis method based on fuzzy logic according to claim 1, characterized in that: In S7, the error condition of the current data is determined based on the error state value after defuzzification in S6, and an error analysis report is output.

9. A computer error analysis device based on fuzzy logic, comprising a computer error analysis method based on fuzzy logic according to any one of claims 1 to 8, characterized in that: include: Data acquisition module: used to obtain basic data; Analysis module: cleans and integrates data; Data fuzzification processing module: convert data into fuzzy variables. Data reasoning module: use fuzzy algorithm to reason about fuzzy edge beams and obtain fuzzy sets. Data conversion module, converts fuzzy sets into precise values, The result output module analyzes and evaluates the errors and outputs error analysis reports.

Images