Intelligent analysis method, system, equipment and medium for fire coal test data of thermal power plant
By using nonparametric rank correlation analysis and time-dependent models, the problems of correlation misjudgment and anomaly removal in the analysis of coal calorific value in thermal power plants were solved, enabling accurate prediction of coal calorific value and adaptive control of equipment, thereby improving the analysis accuracy of coal test data and the reliability of equipment execution.
Patent Information
- Application Number
- CN202510991681.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-18
- Publication Date
- 2025-11-21
AI Technical Summary
Existing technologies suffer from correlation misjudgment, incompatibility with the anomaly removal mechanism of the Chauvin criterion in the analysis of the calorific value of coal in thermal power plants, and lack the ability to adaptively generate storage thresholds based on reproducibility critical values, which makes it impossible for sample preparation equipment to execute classification and storage strategies.
Nonparametric rank correlation analysis was used to quantify the negative correlation between calorific value difference and retest interval time, a time-dependent model of calorific value decay was constructed, the coefficients of the prediction equation were dynamically corrected, and a differentiating storage time threshold instruction for coal type was generated based on the deviation of calorific value from the critical value.
The negative correlation between calorific value decay and time span is accurately quantified, and a calorific value decay model is dynamically constructed to achieve adaptive prediction for high volatile coal types, thereby improving the reliability of long-term prediction and the closed-loop decision-making capability of equipment execution.
Smart Images
Figure CN120998322A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of intelligent monitoring and control of coal quality, and particularly relates to an intelligent analysis method, system, device and medium for coal test data of a thermal power plant. BACKGROUND
[0002] Coal calorific value analysis is a core link of fuel management in a thermal power plant, and its accuracy directly affects the calculation of unit thermal efficiency, carbon emission and cost control. The current mainstream technology relies on traditional statistical tools to process test data, and establishes a calorific value prediction model through linear regression or an empirical formula. In recent years, with the popularization of industrial big data technology, some research attempts to introduce machine learning algorithms to optimize prediction accuracy, and to monitor data anomalies using a quality control chart. These methods have improved the prediction ability of a single scenario, but generally rely on the normal distribution assumption of data, and have not systematically quantified the time-varying influence of coal oxidation characteristics on calorific value decay.
[0003] Traditional parametric statistical methods require data to conform to a normal distribution, while actual coal retest data often has outliers and skewed distribution. Forced application of such methods will lead to correlation misjudgment, and cannot be compatible with the abnormal elimination mechanism of Shewhart criterion. Existing models ignore the differentiated influence of coalification degree on decay rate, resulting in a prediction equation that cannot dynamically respond to coal switching. Current technology can draw a quality control chart, but does not establish a linkage mechanism with prediction model correction, and lacks the ability to adaptively generate a save threshold based on a reproducibility critical value, making it impossible for sample preparation equipment to perform a classified storage strategy. SUMMARY
[0004] In view of the above existing problems, the present application provides an intelligent analysis method, system, device and medium for coal test data of a thermal power plant, to solve the problems of correlation misjudgment, incompatibility with the abnormal elimination mechanism of Shewhart criterion, inability to dynamically respond to coal switching, and lack of adaptive generation of a save threshold based on a reproducibility critical value, which makes it impossible for sample preparation equipment to perform a classified storage strategy.
[0005] To solve the above technical problems, an intelligent analysis method for coal test data of a thermal power plant is proposed, which includes,
[0006] Collecting test data of coal retest samples, performing non-parametric rank correlation analysis on non-normal distribution data, quantifying the negative correlation strength of calorific value difference and retest interval time, identifying the monotonic association between calorific value difference and retest interval time; based on the negative correlation strength and associated coal oxidation sensitivity parameters, constructing a time-dependent model of calorific value decay, and predicting the calorific value change using linear regression; comparing the calorific value-time quality control chart of a standard coal sample and a coal sample to be tested, and dynamically correcting the coefficients of the prediction equation.
[0007] As a preferred scheme of the intelligent analysis method for coal-fired power plant coal combustion test data provided by the application, the test data of the coal combustion retest sample comprises a heat value difference of the coal combustion retest sample, a retest interval time and a coal type.
[0008] As a preferred scheme of the intelligent analysis method for coal-fired power plant coal combustion test data provided by the application, the quantification of the negative correlation strength between the heat value difference and the retest interval time comprises performing non-parametric rank correlation analysis on non-normal distribution data to quantize the negative correlation strength between the heat value difference and the retest interval time, and identifying a monotonic association relationship between the heat value difference and the retest interval time.
[0009] The identification of the monotonic association relationship between the heat value difference and the retest interval time comprises drawing a scatter plot of the heat value difference and the retest interval time, identifying a monotonic decreasing trend, arranging the two groups of data in descending order of numerical value respectively and assigning rank numbers, taking average rank numbers of equivalent data, calculating a square sum of rank number differences of each pair of data points, generating a correlation coefficient, and constructing a normal distribution statistic for two-sided hypothesis testing.
[0010] As a preferred scheme of the intelligent analysis method for coal-fired power plant coal combustion test data provided by the application, the construction of the time-dependent model of heat value decay comprises introducing a coal sample oxidation characteristic parameter, and dynamically adjusting the heat value decay rate of the coal sample in storage based on the coalification degree of the coal type.
[0011] As a preferred scheme of the intelligent analysis method for coal-fired power plant coal combustion test data provided by the application, the prediction of the heat value change amount comprises selecting more than or equal to 5 groups of coal samples, retesting the heat values at a pre-set storage time node, removing abnormal data caused by environmental interference by using the Chauvenet criterion, and solving linear regression parameters to predict the heat value change amount.
[0012] The Chauvenet criterion is |ΔQ|>2σ, wherein σ is a sample standard deviation.
[0013] The formula for solving the linear regression parameters to predict the heat value change amount is expressed as:
[0014] ΔQ=k·t+b
[0015] Wherein, ΔQ is a retest heat value difference of the coal sample, t is a storage time of the coal sample, k is a linear regression slope, i.e., a heat value decay rate, and b is a linear regression intercept.
[0016] As a preferred scheme of the intelligent analysis method for coal-fired power plant coal combustion test data provided by the application, the dynamic correction of the prediction equation coefficients comprises drawing a heat value-time quality control comparison graph of a standard coal sample and a coal sample to be tested, and correcting the linear regression coefficients according to the differences in the comparison graph.
[0017] The heat value-time quality control comparison chart of the standard coal sample and the to-be-tested coal sample is used to identify the heat value fluctuation range of the standard coal sample, the monotone decreasing trend of the heat value of the to-be-tested coal sample containing lignite and long flame coal, and the critical value of the heat value deviation reproducibility when the coal sample is stored for more than 30 days.
[0018] As a preferred scheme of the intelligent analysis method of coal-fired power plant coal test data, the dynamic correction of the prediction equation coefficient further comprises generating a coal sample storage duration threshold instruction based on the result of the heat value deviation reproducibility critical value.
[0019] The coal sample storage duration threshold comprises setting a first threshold value for a coal type with high coalification degree and setting a second threshold value for a coal type with low coalification degree, and the second threshold value is smaller than the first threshold value; wherein the first threshold value is 30 days, and the second threshold value is 15 days.
[0020] As a preferred scheme of the intelligent analysis system of coal-fired power plant coal test data, the system comprises a data acquisition and preprocessing module, a statistical correlation analysis module, a heat value attenuation dynamic modeling module, and a quality control and decision optimization module.
[0021] The data acquisition and preprocessing module is used to collect original data in the coal test process of the coal-fired power plant, and to construct a data set through threshold filtering and standardization processing.
[0022] The statistical correlation analysis module is used to adopt a non-parametric rank correlation analysis method, to generate a correlation coefficient by calculating a rank difference sum of squares, and to verify the significant negative correlation between the heat value difference and the retest interval time by hypothesis testing.
[0023] The heat value attenuation dynamic modeling module is used to construct an adaptive heat value attenuation model based on the statistical correlation results and the coal type characteristics, to define a coal type sensitivity coefficient, to calculate an attenuation rate, and to fit a linear equation through multi-batch storage tests.
[0024] The quality control and decision optimization module is used to draw a heat value-time curve, to identify an inflection point, to dynamically correct a prediction equation coefficient, and to generate a coal type classification storage threshold value and an equipment control instruction.
[0025] A computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the intelligent analysis method of coal-fired power plant coal test data when executing the computer program.
[0026] A computer readable storage medium stores a computer program, and the computer program implements the steps of the intelligent analysis method of coal-fired power plant coal test data when executed by a processor.
[0027] The beneficial effects of the present application: the present application collects the heat value difference of the coal combustion retest sample, the retest interval time and the coal type, adopts a non-parametric statistical method independent of data distribution assumption, accurately quantifies the negative correlation strength of heat value decay and time span, effectively overcomes the failure risk of traditional parameter statistics in skewed data; and innovatively introduces the coal oxidation sensitivity parameter, dynamically constructs the time-dependent model of heat value decay, realizes the adaptive prediction of the accelerated oxidation characteristics of high volatile coal; by comparing the heat value-time variation law of the standard coal sample and the coal sample to be measured, the stable fluctuation interval of bituminous coal, the monotonic decay inflection point of the easily oxidized coal and the critical deviation characteristics of the over-storage are identified, and the prediction model parameters are corrected in real time according to the above, which significantly improves the reliability of long-term prediction; and based on the identification result of the heat value deviation from the critical value, the differentiated storage time threshold instruction of the coal is automatically generated, which drives the sample preparation system to perform priority processing and over-period control on the high sensitivity coal, forming a closed-loop decision chain from data analysis to equipment execution. BRIEF DESCRIPTION OF DRAWINGS
[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0029] Figure 1 The overall flowchart of the intelligent analysis method of the coal-fired power plant coal test data provided by an embodiment of the present application.
[0030] Figure 2 The intelligent analysis method of the coal-fired power plant coal test data provided by an embodiment of the present application.
[0031] Figure 3 The system scheme flowchart of the intelligent analysis system of the coal-fired power plant coal test data provided by an embodiment of the present application. DETAILED DESCRIPTION
[0032] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the protection scope of the present application.
[0033] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details set forth in this description, that the present application can be practiced with only some of the details set forth in this description, that the present application can be practiced with other elements in addition to or in place of those set forth in this description. In other instances, well known structures and devices are shown in block diagram form in order to not interfere with the understanding of the present application.
[0034] Second, the "one embodiment" or "an embodiment" referred to herein means a specific feature, structure, or characteristic under discussion. The appearance of "in one embodiment" at various places in this specification does not necessarily refer to the same embodiment, nor does it necessarily refer to a single embodiment. Rather, the various embodiments described for the present application are combinable and interchangeable at all levels.
[0035] The present application is described in detail below with reference to the attached drawing figures, wherein the components in the drawings are not necessarily to scale, and the same reference numerals that identify similar components are used throughout several figures. Moreover, certain prior art can also be described and need not be set forth in this description itself.
[0036] Meanwhile, in the description of the present application, it should be noted that the terms "upper and lower, inner and outer" and the like indicate the positional or directional relationship based on the positional or directional relationship shown in the drawings, and are merely for the purpose of facilitating the description of the present application and simplifying the description, and therefore should not be construed as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore should not be construed as limiting the present application. In addition, the terms "first, second or third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0037] In the present application, unless otherwise explicitly specified and limited, the terms "mounting, connection, and connection" should be understood broadly, for example: it can be fixed connection, detachable connection or integral connection; it can also be mechanical connection, electrical connection or direct connection, it can also be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.
[0038] Embodiment 1, refer to Figure 1 and Figure 2 The first embodiment of the present application provides an intelligent analysis method for coal-fired power plant coal test data, which comprises:
[0039] S1: Collecting test data of coal retest samples, performing non-parametric rank correlation analysis on non-normal distribution data, and quantifying the negative correlation strength of heat value difference and retest interval time.
[0040] Further, the collection of test data of coal retest samples includes the collection of heat value difference, retest interval time and coal type of coal retest samples.
[0041] The quantification of the negative correlation intensity between the heat value difference and the retest interval time includes performing non-parametric rank correlation analysis on non-normal distribution data to quantify the negative correlation intensity between the heat value difference and the retest interval time, and identifying a monotonic correlation between the heat value difference and the retest interval time.
[0042] The identification of the monotonic correlation between the heat value difference and the retest interval time includes drawing a scatter plot of the heat value difference and the retest interval time, identifying a monotonic decreasing trend, arranging two groups of data in descending order of values and assigning rank numbers, taking the average rank number of equivalent data, calculating the square sum of rank number differences of each pair of data points, generating a correlation coefficient, and constructing a normal distribution statistic for two-sided hypothesis testing.
[0043] The formula for calculating the rank number difference of each pair of data is:
[0044] d i =R T,i -R ΔQ,i
[0045] Where d i is the rank number difference of the i-th pair of data points, R t,i is the rank number of the retest interval time T of the i-th sample, and R ΔQ,i is the rank number of the heat value difference AQ of the i-th sample.
[0046] The formula for calculating the Spearman correlation coefficient is:
[0047]
[0048] Where r s is the Spearman rank correlation coefficient, d i is the rank number difference of the i-th pair of data points, n is the number of effective samples, and i is the variable index.
[0049] In the embodiments of the present application, the non-parametric rank correlation analysis includes quantifying the negative correlation between the heat value difference and the time through the Spearman rank correlation.
[0050] In an alternative embodiment, the non-parametric rank correlation analysis includes pairing the retest interval time and the heat value difference sample data to form data pairs, comparing the rank number consistency of all data pairs, and counting the number of harmonious pairs and the number of inharmonious pairs, calculating the Kendall coefficient, and evaluating the significance through the permutation test.
[0051] In another alternative embodiment, the non-parametric rank correlation analysis includes calculating the sign of each pair of data, counting the proportion of positive signs, and comparing it with the null hypothesis, using the binomial test to determine the significance, and applying it to the embedded equipment deployment and rapid retest decision-making scenarios in the sample making workshop.
[0052] Further, constructing the normal distribution statistics includes constructing the normal distribution statistics and calculating a two-sided test P value, which is expressed by the formula:
[0053]
[0054] P = 2 x (1 - normcdf (X))
[0055] wherein X is the normalized test statistics for converting the Spearman correlation coefficient into a standard normal distribution, r s is the Spearman rank correlation coefficient, n is the effective sample size, P is the significance level, normcdf (X) is the cumulative distribution function of the standard normal distribution, and the probability area on the left side of X is calculated; when P < 0.05, it is judged that there is a strong correlation.
[0056] It should be noted that by collecting the heat value difference of the coal combustion retest sample, the retest interval time and the coal type, using a non-parametric statistical method independent of the data distribution assumption, the negative correlation strength between the heat value decay and the time span is accurately quantified, the failure risk of the traditional parameter statistics in the skewed data is effectively overcome, the coal oxidation sensitivity parameter is innovatively introduced, the time-dependent model of heat value decay is dynamically constructed, and adaptive prediction of the accelerated oxidation characteristics of high volatile coal is realized.
[0057] S2: Based on the negative correlation strength, and in association with the coal oxidation sensitivity parameter, a time-dependent model of heat value decay is constructed, and linear regression is used to predict the heat value change.
[0058] Further, the time-dependent model of heat value decay includes introducing the coal sample oxidation characteristic parameter, and the parameter is dynamically adjusted based on the heat value decay rate of coal with low coalification degree in storage.
[0059] The coal oxidation sensitivity parameter is dynamically set according to the dry basis volatile matter of the coal: a high sensitivity coefficient is given to coal with volatile matter ≥ 30%, and a low sensitivity coefficient is given to coal with volatile matter < 30%, which is expressed by the formula:
[0060]
[0061] wherein a is the oxidation sensitivity coefficient, V daf is the dry ash-free basis volatile matter of the coal sample.
[0062] It should be noted that the time-dependent model is expressed as:
[0063] γ = a x β x t
[0064] wherein γ is the heat value decay rate, β is the basic decay rate derived from the Spearman correlation coefficient, t is the storage time of the coal sample, and a is the oxidation sensitivity coefficient.
[0065] In the embodiments of the present application, the oxidation sensitivity of coal types includes dividing the sensitivity coefficient according to the volatile matter threshold (30%).
[0066] In an alternative embodiment, the oxidation sensitivity of coal types includes retesting the calorific value of each coal sample every day for 5 days before storage, calculating the daily average decay rate, dividing the sensitivity coefficient, and applying it to multi-coal source blending power plants and coal quality rapid detection scenarios.
[0067] In another alternative embodiment, the oxidation sensitivity of coal types includes determining the maceral of the coal sample by vitrinite reflectance, and when the vitrinite content is greater than 60% and the inertinite is less than 20%, it is determined as a high-activity coal type, otherwise, the volatile matter threshold is used for secondary classification, which is applied to high-precision fuel management power plants and coal types with similar volatile matter but large differences in oxidation characteristics.
[0068] Further, the predicted calorific value change amount includes selecting more than or equal to 5 groups of coal samples, retesting the calorific value at the preset storage time node, using the Chauvenet criterion to remove abnormal data caused by environmental interference, and calculating the linear regression parameter to predict the calorific value change amount;
[0069] The Chauvenet criterion is |ΔQ|>2σ, where σ is the sample standard deviation;
[0070] The formula for calculating the linear regression parameter to predict the calorific value change amount is represented as:
[0071] ΔQ=k·t+b
[0072] Where ΔQ is the difference in retested calorific value of the coal sample, t is the storage time of the coal sample, k is the linear regression slope, i.e. the calorific value decay rate, and b is the linear regression intercept.
[0073] It should be noted that by comparing the calorific value-time variation law of the standard coal sample and the coal sample to be tested, the stable fluctuation interval of bituminous coal, the monotonic decay inflection point of the easily oxidized coal type, and the critical deviation characteristics of the over-storage are identified, and the prediction model parameters are corrected in real time, which significantly improves the reliability of long-term prediction.
[0074] S3: Comparing the calorific value-time quality control chart of the standard coal sample and the coal sample to be tested, and dynamically correcting the coefficients of the prediction equation.
[0075] Further, the dynamic correction of the prediction equation coefficients includes drawing the calorific value-time quality control comparison chart of the standard coal sample and the coal sample to be tested, and correcting the linear regression coefficients according to the difference in the comparison chart.
[0076] Using the calorific value-time quality control comparison chart of the standard coal sample and the coal sample to be tested, the calorific value fluctuation range of the standard coal sample of bituminous coal, the monotonic downward trend of the coal sample to be tested containing lignite and long flame coal, and the identification of the calorific value deviation reproducibility critical value when the coal sample is stored for more than 30 days.
[0077] It should be noted that based on the identification result of the heat value deviating from the critical value, the coal type differentiated storage duration threshold instruction is automatically generated, the sample preparation system is driven to perform preferential processing and overage control on the high sensitivity coal type, and a closed loop decision chain from data analysis to equipment execution is formed.
[0078] Further, the dynamically modified prediction equation coefficient further includes generating a coal sample storage duration threshold instruction based on the result of the heat value deviating from the reproducibility critical value;
[0079] The coal sample storage duration threshold includes setting a first threshold value for coal types with high coalification degree and setting a second threshold value for coal types with low coalification degree, and the second threshold value is less than the first threshold value; wherein the first threshold value is 30 days and the second threshold value is 15 days.
[0080] The correction condition and rule formula are represented as:
[0081] When t is greater than or equal to 15 and |ΔQ| is greater than 0.12:
[0082]
[0083] When t is greater than or equal to 30 and |ΔQ| is greater than 0.3: k new = 1.2k
[0084] Wherein, k new is the corrected heat value decay rate slope, k is the heat value decay rate, |ΔQ 实测 | is the absolute value of the actual retested heat value difference, |ΔQ 预测 | is the absolute value of the model predicted heat value difference, and ΔQ is the coal sample retested heat value difference.
[0085] In the embodiments of the present application, the dynamically modified prediction equation coefficient includes manually identifying the inflection point correction coefficient by comparing the standard coal sample and the to-be-tested coal sample curve.
[0086] In an alternative embodiment, the dynamically modified prediction equation coefficient includes first-order difference of the to-be-tested coal sample heat value-time sequence, when the first-order difference of three consecutive time points is less than 0.05, it is determined as a decay inflection point, the data weight after the inflection point is improved to 80% to re-fit the regression equation, and it is applied in unattended intelligent laboratory and high-frequency retest scene.
[0087] In another alternative embodiment, the dynamically modified prediction equation coefficient includes training an LSTM network to learn historical quality control chart error patterns, when the real-time error is greater than 0.1, the LSTM output compensation value is called to modify the prediction equation, and it is applied in coastal high temperature and high humidity environment power plant and large group power plant equipped with AI platform.
[0088] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.
[0089] Embodiment 2, refer to Figure 2 For the second embodiment of the present application, an intelligent analysis method for coal-fired power plant coal assay data is provided. In order to verify the beneficial effects of the present application, scientific demonstration is carried out through experiments.
[0090] Considering that the data does not conform to the normal distribution and there are outliers, Spearman is selected to analyze the correlation.
[0091] First, use the scatter plot to judge the monotonicity, as shown in Figure 2 The retest sample heat value difference and the retest interval time have a certain monotonic relationship, and the absolute value of the retest sample heat value difference increases with the increase of the retest interval time, and the retest sample heat value difference and the interval days have a negative correlation.
[0092] Second, calculate the correlation, sort the original data in descending order, and give the grade, when the values of two data are equal, calculate the average value of the value grade as the grade number, calculate the grade difference of two groups of data and the Spearman correlation coefficient, as shown in Table 1:
[0093] Table 1 Repeated sample heat value difference and interval days sorting table
[0094]
[0095]
[0096] Third, hypothesis test, because the sample number n is greater than 30, it belongs to large sample situation, so the statistical quantity conforming to normal distribution is constructed, in matlab, bilateral test is used, the retest sample heat value difference and the retest interval time have strong negative correlation, that is, the longer the retest time interval, the more negative the retest sample heat value difference.
[0097] Embodiment 3, refer to Figure 3 For the third embodiment of the present application, the embodiment provides an intelligent analysis system for coal-fired power plant coal assay data, which includes a data acquisition and preprocessing module, a statistical correlation analysis module, a heat value decay dynamic modeling module, and a quality control and decision optimization module.
[0098] The data acquisition and preprocessing module is configured to collect original data in a coal combustion test of a thermal power plant, and to construct a data set through threshold filtering and standardization processing.
[0099] The statistical correlation analysis module is configured to use a non-parametric rank correlation analysis method, to generate a correlation coefficient by calculating a rank difference sum of squares, and to verify a significant negative correlation between the heat value difference and the retest interval time by using a hypothesis test.
[0100] The heat value attenuation dynamic modeling module is configured to construct an adaptive heat value attenuation model based on statistical correlation results and coal type characteristics, to define a coal type sensitive coefficient, to calculate an attenuation rate, and to fit a linear equation through multi-batch preservation tests.
[0101] The quality control and decision optimization module is configured to draw a heat value-time curve, to identify an inflection point, to dynamically correct a prediction equation coefficient, to generate a coal type classification preservation threshold, and to generate an equipment control instruction.
[0102] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.
[0103] Embodiment 4, which is different from the first three embodiments, is as follows:
[0104] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the present application that essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.
[0105] The logic and / or steps represented in the flow diagrams and / or otherwise described herein, for example, can be considered as a sequence of executable instructions, and can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. Just by way of example, a computer-readable medium can be any device or apparatus that can store and convey instructions for execution by the instruction execution system, apparatus, or device. With respect to the present description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[0106] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via an optical scanner, then compiled, interpreted, or otherwise processed, and stored in a computer memory in a form that can be later executed by a computer.
[0107] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following technologies can be used: a combination of discrete logic circuits having logic gates for implementing logic functions upon an application of data signals, application specific integrated circuits having logic gates, field programmable gate arrays (FPGA), or other components, in any suitable combination.
Claims
1. An intelligent analysis method for coal-fired power plant coal assay data, characterized in that: The application relates to a method for predicting the change of the calorific value of coal samples. The application comprises the following steps: Collecting test data of retest samples of coal, performing non-parametric rank correlation analysis on non-normal distribution data, quantifying the negative correlation strength between the calorific value difference and the retest interval time, and identifying the monotonic correlation between the calorific value difference and the retest interval time; Based on the negative correlation strength and in association with the oxidation sensitivity parameter of the coal type, a time-dependent model of calorific value attenuation is constructed, and the linear regression is used to predict the calorific value change; 2. The intelligent analysis method of coal-fired power plant coal assay data according to claim 1, characterized in that: Comparing the calorific value-time quality control chart of the standard coal sample and the coal sample to be measured, and dynamically correcting the coefficients of the prediction equation.
3. The intelligent analysis method of coal-fired power plant coal assay data according to claim 2, characterized in that: The collection of test data of retest samples of coal comprises the following steps: collecting the calorific value difference, the retest interval time and the coal type of the retest samples of coal. The quantification of the negative correlation strength between the calorific value difference and the retest interval time comprises the following steps: performing non-parametric rank correlation analysis on non-normal distribution data, quantifying the negative correlation strength between the calorific value difference and the retest interval time, and identifying the monotonic correlation between the calorific value difference and the retest interval time.
4. The intelligent analysis method of coal-fired power plant coal assay data according to claim 3, characterized in that: The identification of the monotonic correlation between the calorific value difference and the retest interval time comprises the following steps: drawing a scatter plot of the calorific value difference and the retest interval time, identifying the monotonic decreasing trend, arranging the two groups of data in descending order of value and assigning rank numbers, taking the average rank number of the equivalent data, calculating the square sum of the rank number difference of each pair of data points, generating the correlation coefficient, and constructing the normal distribution statistic for two-sided hypothesis testing.
5. The intelligent analysis method of coal-fired power plant combustion test data according to claim 4, characterized in that: The construction of the time-dependent model of calorific value attenuation comprises the following steps: introducing the oxidation characteristic parameter of the coal sample, and dynamically adjusting the parameter based on the calorific value attenuation rate of the coal type with low coalification degree in storage. The prediction of the calorific value change comprises the following steps: selecting more than five groups of coal samples, retesting the calorific value at the preset storage time node, removing abnormal data caused by environmental interference by adopting the Chauvenet criterion, and calculating the linear regression parameter to predict the calorific value change. The Chauvenet criterion is: |Delta Q|>2sigma, wherein sigma is the sample standard deviation. The formula for calculating the linear regression parameter to predict the calorific value change is expressed as: Delta Q=k.t+b 6. The intelligent analysis method of coal-fired power plant coal assay data according to claim 5, characterized in that: Wherein, Delta Q is the calorific value difference of the coal sample, t is the storage time of the coal sample, k is the linear regression slope, that is, the calorific value attenuation rate, and b is the linear regression intercept. The dynamic correction of the coefficients of the prediction equation comprises the following steps: drawing a calorific value-time quality control comparison chart of the standard coal sample and the coal sample to be measured, and correcting the linear regression coefficients according to the differences in the comparison chart.
7. The intelligent analysis method of coal-fired power plant coal assay data according to claim 6, characterized in that: The calorific value-time quality control comparison chart of the standard coal sample and the coal sample to be measured is used to identify the calorific value fluctuation range of the standard bituminous coal sample, the monotonic decreasing trend of the calorific value of the brown coal and long flame coal sample to be measured, and the calorific value deviation reproducibility critical value when the coal sample is stored for more than 30 days. The dynamic correction of the coefficients of the prediction equation further comprises the following steps: based on the result of the calorific value deviation reproducibility critical value, a coal sample storage time threshold instruction is generated.
8. A system for intelligent analysis of coal combustion test data of a thermal power plant using the method as claimed in any one of claims 1 to 7, characterized in that: The coal sample storage time threshold comprises the following steps: setting a first threshold value for the coal type with high coalification degree, and setting a second threshold value for the coal type with low coalification degree, and the second threshold value is smaller than the first threshold value; wherein the first threshold value is 30 days, and the second threshold value is 15 days. The method comprises the following modules: a data collection and preprocessing module, a statistical correlation analysis module, a calorific value attenuation dynamic modeling module, and a quality control and decision optimization module. The data acquisition and preprocessing module is used for collecting original data in the coal test process of the thermal power plant, and constructing a data set through threshold filtering and standardization processing; The statistical correlation analysis module is used for adopting a non-parametric rank correlation analysis method, generating a correlation coefficient by calculating a rank difference sum of squares, and verifying a significant negative correlation between the heat value difference and the retest interval time by using a hypothesis test; The heat value attenuation dynamic modeling module is used for constructing an adaptive heat value attenuation model based on the statistical correlation result and the coal type characteristics, defining a coal type sensitive coefficient, calculating an attenuation rate, and fitting a linear equation through multi-batch saving tests; The quality control and decision optimization module is used for drawing a heat value-time curve, identifying an inflection point, dynamically correcting a prediction equation coefficient, generating a coal type classification saving threshold, and generating an equipment control instruction. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the intelligent analysis method of the coal test data of the thermal power plant according to any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the intelligent analysis method of the coal test data of the thermal power plant according to any one of claims 1 to 7.