CN110991660A - Situation analysis method, system and storage medium of LSSVM-ARIMA model based on locust optimization - Google Patents
Situation analysis method, system and storage medium of LSSVM-ARIMA model based on locust optimization Download PDFInfo
- Publication number
- CN110991660A CN110991660A CN201910592467.2A CN201910592467A CN110991660A CN 110991660 A CN110991660 A CN 110991660A CN 201910592467 A CN201910592467 A CN 201910592467A CN 110991660 A CN110991660 A CN 110991660A
- Authority
- CN
- China
- Prior art keywords
- lssvm
- locust
- optimization
- arima model
- analysis method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a situation analysis method, a system and a storage medium of an LSSVM-ARIMA model based on locust optimization, belonging to the field of machine learning and data mining and characterized by comprising the following steps: (1) randomly initializing initial position of locust groupAnd(2) determining an objective function(3) Carrying out position updating; (4) repeating the steps (1) and (2) and outputting c and sigma; (5) establishing LSSVM model andan ARIMA model; calculating the predicted result y1(t); (6) determining the Low frequency component Aj(t) and a high-frequency component Dj(t); (7) obtaining a first prediction result y1(t); (8) calculating the predicted result y2(t); (9) will predict the result y1(t) and y2(t) fitting to obtain a final situation result y (t). The invention adopts a mode of combining LSSVM and ARIMA, and realizes parameter optimization on the time-varying potential model by using the locust optimization algorithm. Experimental results show that the enterprise safety production situation prediction method established by the invention is effective, and a reliable method is provided for enterprise safety production management situation analysis.
Description
Technical Field
The invention relates to the technical field of machine learning and data mining, in particular to a situation analysis method, a situation analysis system and a storage medium of an LSSVM-ARIMA model based on locust optimization.
Background
At present, most models can fit original data to a high degree aiming at the situation prediction problem, but the generalization capability of the models is poor. For the situation prediction problem, because the data is a time sequence and changes in real time, the models often show good fitting effect on historical data, but in some newly-appeared data, the prediction capability is greatly reduced. Although the neural network has good generalization capability and memory capability, the convergence rate is too low in the model training process, so that the training time is too long, and the requirement of numerical prediction on timeliness cannot be met. Taking the recurrent neural network RNN as an example, each computation result of the hidden layer of the RNN is related to the current input and the previous result of the hidden layer, and has the capability of memorizing historical data. However, if long-term memory needs to be realized, the current implicit state calculation needs to be hooked with the previous n times of calculation, so that the calculation amount grows exponentially, and the time for model training is greatly increased. Similarly, when the BP neural network adjusts the weights among the neurons in each layer by using an error back propagation method, multiple iterations are often required, which increases the training time of the model and is easy to fall into local optimization during the training process.
At present, with the improvement of the safety production consciousness of enterprises and the exponential growth of mass data, higher requirements are put forward on the feature selection and the prediction performance index of the model. Therefore, a prediction method capable of accurately and efficiently predicting the numerical value change trend in some fields to meet the situation prediction requirement is needed.
Disclosure of Invention
The embodiment of the invention aims to provide a situation analysis method, a system and a storage medium of an LSSVM-ARIMA model based on locust optimization, so that the model can give out the change of the situation according to the situation result and historical data, finally realize the judgment of the safety production management situation of a target object, and effectively improve the supervision efficiency of safety production.
The invention provides a situation analysis method of an LSSVM-ARIMA model based on locust optimization, which comprises the following steps:
(1) parameters α in LSSVM modelkAll values of b are taken as locusts, and the initial positions of the locusts are initialized randomlyAndmaximum number of iterations TmaxMaximum and minimum values c of the parameter cmax、cminWherein αkA lagrange multiplier, b is an offset, and a parameter c is used for avoiding falling into local optimization due to too fast approaching a target value;
(3) Updating the position of the locust swarm searching individual according to the target function;
(4) the updated position is substituted into the step (2) again, the step (2) and the step (3) are repeated, and finally α is outputkB, optimal solution;
(5) according to said αkB, establishing an LSSVM model by the optimal solution; establishing an ARIMA model;
(6) determining a low frequency component A of a time series of safety production management situation dataj(t) and a high-frequency component Dj(t);
(7) The low-frequency component Aj(t) substituting the LSSVM model to obtain a first prediction result y1(t);
(8) Determining different high-frequency components Dj(t) autocorrelation coefficients and partial autocorrelation coefficients p, q; the high frequency component Dj(t) substituting the autocorrelation coefficient and the partial autocorrelation coefficients p and q into the ARIMA model to obtain a second prediction result y2(t);
(9) Predicting the result y of the two parts1(t) and y2(t) fitting to obtain a final situation result y (t).
Optionally, in the situation analysis method of the LSSVM-ARIMA model based on locust optimization, the objective function is determined in the following manner in the step (2)
In the formula, f and l respectively represent an attraction strength parameter and an attraction scale parameter, r represents a comfortable distance, and s represents an influence function of the interaction force of other locusts on the locusts; ubd、lbdRespectively the upper and lower limits of the ith locust on the d-dimensional variable;is a coefficient of linear decreasing, t represents the current iteration number;is the target position of the locust colony,represents the unit vector from the i th locust to the j th locust, xj(t) represents the position of the jth locust in the locust group, xi(t) indicates the position of the ith locust in the locust group, dijIndicating the distance between the two.
Optionally, in the situation analysis method of the LSSVM-ARIMA model based on locust optimization, in step (3), the position of the locust swarm searching individual is updated by the following method:
p is the position transition probability, α, β are the heuristic and desired heuristic, η, respectivelyeIn order to select the desire of a certain location,it is shown that the initial position is,represents the position increment at t +1 cycles.
Optionally, in the situation analysis method of the LSSVM-ARIMA model based on locust optimization, in step (5):
the LSSVM model is as follows:
in the formula (I), the compound is shown in the specification,a nonlinear spatial mapping function;
the ARIMA model is as follows:
wherein the content of the first and second substances,is an autoregressive coefficient, θi(i ═ 1, 2.. times, q) is a moving average coefficient, u is a moving average coefficienttIs an error term.
Optionally, in the situation analysis method of the LSSVM-ARIMA model based on locust optimization, in step (6):
determining a low frequency component A of a time series of safety production management situation dataj(t) and a high-frequency component Dj(t):
aj+1=h0*aj
dj+1=h1*dj
Wherein j is 0, 10As a low-pass decomposition filter, h1For high-pass decomposition filters, ajIs a low frequency coefficient, djIs a high frequency coefficient;
Aj(t)=g0*aj
Dj(t)=g1*dj
in the formula, g0For low-pass reconstruction filters, g1A high pass reconstruction filter.
Optionally, in the situation analysis method of the LSSVM-ARIMA model based on locust optimization, the first prediction result y obtained in the step (7)1(t) is:
optionally, in the situation analysis method of the LSSVM-ARIMA model based on locust optimization, the autocorrelation coefficients and the partial autocorrelation coefficients p and q are obtained in the step (8) as follows:
where k is the hysteresis order, Dj(t) is a high frequency component.
Optionally, the situation analysis method of the LSSVM-ARIMA model based on locust optimization, the second prediction result y obtained in the step (8)2(t) is:
the invention also provides a situation analysis system of the LSSVM-ARIMA model based on locust optimization, which comprises at least one processor and at least one memory, wherein program information is stored in the at least one memory, and the at least one processor reads the program information and then executes any one of the situation analysis methods of the LSSVM-ARIMA model based on locust optimization.
The invention also provides a storage medium, wherein the storage medium is stored with program instructions, and a computer reads the program information and then executes any one of the above situation analysis methods based on the locust optimization LSSVM-ARIMA model.
Compared with the prior art, the technical scheme provided by the embodiment of the invention at least has the following beneficial effects:
(1) the situation analysis method, the system and the storage medium of the LSSVM-ARIMA model based on locust optimization provided by the invention adopt a mode of combining the LSSVM model and the ARIMA model, and decompose a time sequence formed by safety production management situation data into a low-frequency part and a high-frequency part by wavelet decomposition. And carrying out situation analysis on the low-frequency part by adopting an LSSVM model, and carrying out situation analysis on the high-frequency part by adopting an ARIMA model. And finally, fitting the results of the two parts, fully considering the details of the situation data and realizing accurate situation prediction.
(2) The situation analysis method, the system and the storage medium of the LSSVM-ARIMA model based on locust optimization provided by the invention are used for carrying out iterative optimization on model parameters through the steps (2) and (3) so as to realize global optimization of the model parameters.
(3) According to the situation analysis method, system and storage medium of the LSSVM-ARIMA model based on locust optimization, provided by the invention, multiple groups of data are tested, the test result is relatively stable, the generalization capability of the model is improved on the basis of ensuring the prediction precision, and the numerical prediction task can be better completed.
Drawings
FIG. 1 is a flow chart of a situation analysis method of an LSSVM-ARIMA model based on locust optimization according to an embodiment of the present invention;
FIG. 2 is a flow chart of a situation analysis method of an LSSVM-ARIMA model based on locust optimization according to another embodiment of the present invention;
FIG. 3 is a time series line graph of index data;
FIG. 4 is a schematic diagram showing comparison of results of multiple sets of simulation experiments performed on the present invention.
Detailed Description
The embodiments of the present invention will be further described with reference to the accompanying drawings. In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description of the present invention, and do not indicate or imply that the device or assembly referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Wherein the terms "first position" and "second position" are two different positions.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; the two components can be directly connected or indirectly connected through an intermediate medium, and the two components can be communicated with each other. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Taking all the enterprise safety production data in Beijing as an example, the selected data comprises data in aspects of safety production responsibility implementation, hidden danger elimination condition, accident prevention and control capability and the like, and the total number of the data is 322 ten thousand, wherein 311 ten thousand hidden danger data are recorded. The index data shown in fig. 3 is used as an original time series data set, a part of the index data is marked as a training set, a part of the index data is marked as a test set, and the total number of the index data is 221 ten thousand training sample data and 90 ten thousand test sample data. Firstly, carrying out global optimization on parameters of an LSSVM model by using a locust optimization algorithm to find out an optimal solution of the parameters so as to construct the LSSVM model; secondly, dividing the training sample and the test sample into a low-frequency component and a high-frequency component by using a wavelet decomposition algorithm; then, calculating parameters of ARIMA by using the high-frequency components, and constructing a model; and finally, predicting the low-frequency component and the high-frequency component by respectively adopting an LSSVM model and an ARIMA model, and synthesizing the prediction results of the two parts to obtain a final situation prediction result.
The overall flow of the situation prediction method provided by the present invention is shown in fig. 1 and fig. 2, and comparing fig. 1 and fig. 2, it can be seen that, in the embodiment of the present invention, if there is no dependency relationship between the steps, the order of the steps can be adjusted, and the specific steps are as follows:
(1) random initializationTmax、cmax、cmin(ii) a Wherein the content of the first and second substances,anddenotes the initial position of the locust group, TmaxDenotes the maximum number of iterations, cmax、cminRepresenting the maximum and minimum values of c, respectively, the function of the parameter c is to avoid falling into local optima too fast near the target value, where locust is the parameter α of the LSSVM modelkAnd all values of b, αkIs the lagrange multiplier and b is the bias. In this example, the values of the parameters are:Tmax=1000,cmax=1,cmin=0.0001。
Wherein f and l respectively represent the suction strengthThe number and attraction scale parameters are shown, r represents a comfortable distance, and s represents an influence function of the interaction force of the locust on other locusts; ubd、lbdRespectively the upper and lower limits of the ith locust on the d-dimensional variable;for a linearly decreasing coefficient, t represents the current iteration number.Is the target position of the locust colony,represents the unit vector from the i th locust to the j th locust, xj(t) represents the position of the jth locust in the locust group, xi(t) indicates the position of the ith locust in the locust group, dijRepresents the distance between the two; in this example, f and i have values of f ═ 1.5, l ═ 0.5, and ub, respectivelyd、lbdIs 1 and 0.
(3) Updating the position of the locust colony searching individual to find αkAnd b optimal solution:
p is the position transition probability, α, β are the heuristic and desired heuristic, η, respectivelyeIn order to select the desire of a certain location,it is shown that the initial position is,represents the position increment at t +1 cycles;
(4) the updated position is substituted into the step (2) again, the steps (2) and (3) are repeated, and finally α is outputk、b;
(5) The LSSVM model is established as follows:
in the formula (I), the compound is shown in the specification,a nonlinear spatial mapping function; in this example, take
The ARIMA model was established as follows:
wherein the content of the first and second substances,is an autoregressive coefficient, θi(i ═ 1, 2.., 5.) is a moving average coefficient, utIs an error term.
(6) Determining the Low frequency component Aj(t) and a high-frequency component Dj(t):
The safety production management situation data is an ordered sequence arranged according to time sequence, and the low-frequency component A of the time sequence is firstly determinedj(t) and a high-frequency component Dj(t):
aj+1=h0*aj
di+1=h1*dj;
Wherein j is 0, 10As a low-pass decomposition filter, h1For high-pass decomposition filters, ajIs a low frequency coefficient, djFor high frequency coefficient, the invention adopts mallat algorithm to determine ajAnd dj. In this example, a low-pass decomposition filterHigh-pass decomposition filter
Aj(t)=g0*aj
Dj(t)=g1*dj;
In the formula, g0For low-pass reconstruction filters, g1A high pass reconstruction filter. In this example, the low-pass reconstruction filter and the high-pass reconstruction filter are respectively
(7) Low frequency component Aj(t) substituting the LSSVM model to obtain a prediction result y1(t) is:
(8) determining autoregressive coefficients and moving average coefficients p, q:
where k is the hysteresis order, Dj(t) is a high frequency component. In this example, p and q have values of 2 and 5
Will high frequency component Dj(t) substituting into ARIMA model to obtain the prediction result y2(t) is:
(9) predicting the result y of the two parts1(t) and y2(t) fitting to obtain the final situationResults y (t).
In order to verify the situation prediction precision and the generalization capability of the model, the precedent accuracy rate test experiment of a plurality of groups of situation analysis models is carried out by using the precedent verification data set, and the experimental result is shown in the simulation result schematic diagram shown in the table 1 and is shown in fig. 4.
TABLE 1 test results
Serial number | Accuracy (%) |
1 | 99.70 |
2 | 99.77 |
3 | 99.75 |
4 | 99.82 |
5 | 99.85 |
As can be seen from the simulation result table 1 and FIG. 4, after a plurality of groups of experiments are carried out by using the same verification data set, the accuracy is between 99.7% and 99.9%, and the fluctuation range is only 0.2%, which indicates that the situation prediction method established by the invention has higher stability on the basis of keeping higher accuracy, and can meet the judgment requirement of the target object safety production management situation. The method adopted by the invention is accurate and reliable, and provides a reliable method for analyzing the safety production management situation of the enterprise.
Example 2
The embodiment provides a readable storage medium, wherein the storage medium stores program instructions, and after the program instructions are read by a computer, the computer executes the situation analysis method of the grasshopper optimization-based LSSVM-ARIMA model in the embodiment 1.
Example 3
The embodiment provides a situation analysis system of a grasshopper optimization-based LSSVM-ARIMA model, which comprises at least one processor and at least one memory, wherein program instructions are stored in the at least one memory, and the at least one processor executes the situation analysis method of the grasshopper optimization-based LSSVM-ARIMA model according to the embodiment 1 after reading the program instructions.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A posture analysis method of an LSSVM-ARIMA model based on locust optimization is characterized by comprising the following steps:
(1) parameters α in LSSVM modelkAll values of b are taken as locusts, and the initial positions of the locusts are initialized randomlyAndmaximum number of iterations TmaxMaximum and minimum values c of the parameter cmax、cminWherein αkFor lagrange multipliers, b is the bias, and the parameter c is used to avoid approaching the target valueToo fast to fall into local optima;
(3) Updating the position of the locust swarm searching individual according to the target function;
(4) the updated position is substituted into the step (2) again, the step (2) and the step (3) are repeated, and finally α is outputkB, optimal solution;
(5) according to said αkB, establishing an LSSVM model by the optimal solution; establishing an ARIMA model;
(6) determining a low frequency component A of a time series of safety production management situation dataj(t) and a high-frequency component Dj(t);
(7) The low-frequency component Aj(t) substituting the LSSVM model to obtain a first prediction result y1(t);
(8) Determining different high-frequency components Dj(t) autocorrelation coefficients and partial autocorrelation coefficients p, q; the high frequency component Dj(t) substituting the autocorrelation coefficient and the partial autocorrelation coefficients p and q into the ARIMA model to obtain a second prediction result y2(t);
(9) Predicting the result y of the two parts1(t) and y2(t) fitting to obtain a final situation result y (t).
2. The locustasis optimization-based LSSVM-ARIMA model situation analysis method according to claim 1, wherein the objective function is determined in step (2) by
In the formula, f and l respectively represent an attraction strength parameter and an attraction scale parameter, r represents a comfortable distance, and s represents an influence function of the interaction force of other locusts on the locusts; ubd、lbdRespectively the upper and lower limits of the ith locust on the d-dimensional variable;is a coefficient of linear decreasing, t represents the current iteration number;is the target position of the locust colony,represents the unit vector from the i th locust to the j th locust, xj(t) represents the position of the jth locust in the locust group, xi(t) indicates the position of the ith locust in the locust group, dijIndicating the distance between the two.
3. The posture analysis method of grasshopper optimization-based LSSVM-ARIMA model according to claim 2, characterized in that the position of the grasshopper population searching individuals is updated in step (3) by the following method:
4. The locustasis optimization-based LSSVM-ARIMA model situation analysis method according to claim 3, wherein in step (5):
the LSSVM model is as follows:
in the formula (I), the compound is shown in the specification,a nonlinear spatial mapping function;
the ARIMA model is as follows:
5. The locustasis optimization-based LSSVM-ARIMA model situation analysis method according to claim 4, wherein in step (6):
determining a low frequency component A of a time series of safety production management situation dataj(t) and a high-frequency component Dj(t):
aj+1=h0*aj
dj+1=h1*dj
Wherein j is 0, 10As a low-pass decomposition filter, h1For high-pass decomposition filters, ajIs a low frequency coefficient, djIs a high frequency coefficient;
Aj(t)=g0*aj
Dj(t)=g1*dj
In the formula, g0For low-pass reconstruction filters, g1A high pass reconstruction filter.
7. the posture analysis method of grasshopper optimization-based LSSVM-ARIMA model according to claim 6, characterized in that the autocorrelation coefficients and the partial autocorrelation coefficients p, q are obtained in step (8) by:
where k is the hysteresis order, Dj(t) is a high frequency component.
9. a posture analysis system based on a grasshopper optimization LSSVM-ARIMA model, comprising at least one processor and at least one memory, wherein at least one memory stores program information, and at least one processor executes the posture analysis method based on the grasshopper optimization LSSVM-ARIMA model according to any one of claims 1 to 8 after reading the program information.
10. A storage medium storing program instructions, wherein a computer reads the program information to execute the posture analysis method based on the grasshopper-optimized LSSVM-ARIMA model according to any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910592467.2A CN110991660A (en) | 2019-07-03 | 2019-07-03 | Situation analysis method, system and storage medium of LSSVM-ARIMA model based on locust optimization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910592467.2A CN110991660A (en) | 2019-07-03 | 2019-07-03 | Situation analysis method, system and storage medium of LSSVM-ARIMA model based on locust optimization |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110991660A true CN110991660A (en) | 2020-04-10 |
Family
ID=70081559
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910592467.2A Pending CN110991660A (en) | 2019-07-03 | 2019-07-03 | Situation analysis method, system and storage medium of LSSVM-ARIMA model based on locust optimization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110991660A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111947624A (en) * | 2020-08-12 | 2020-11-17 | 上海卫星工程研究所 | Time-varying situation matrix driven space-based wide-area distributed situation sensing method and system |
CN112766865A (en) * | 2021-03-02 | 2021-05-07 | 河南科技学院 | Internet e-commerce warehousing dynamic scheduling method considering real-time orders |
-
2019
- 2019-07-03 CN CN201910592467.2A patent/CN110991660A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111947624A (en) * | 2020-08-12 | 2020-11-17 | 上海卫星工程研究所 | Time-varying situation matrix driven space-based wide-area distributed situation sensing method and system |
CN111947624B (en) * | 2020-08-12 | 2022-03-18 | 上海卫星工程研究所 | Time-varying situation matrix driven space-based wide-area distributed situation sensing method and system |
CN112766865A (en) * | 2021-03-02 | 2021-05-07 | 河南科技学院 | Internet e-commerce warehousing dynamic scheduling method considering real-time orders |
CN112766865B (en) * | 2021-03-02 | 2023-09-22 | 河南科技学院 | Internet E-commerce warehouse dynamic scheduling method considering real-time order |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111832627B (en) | Image classification model training method, classification method and system for suppressing label noise | |
CN111563706A (en) | Multivariable logistics freight volume prediction method based on LSTM network | |
CN108986470A (en) | The Travel Time Estimation Method of particle swarm algorithm optimization LSTM neural network | |
CN111027732B (en) | Method and system for generating multi-wind power plant output scene | |
CN110659722A (en) | AdaBoost-CBP neural network-based electric vehicle lithium ion battery health state estimation method | |
CN114662780A (en) | Carbon emission prediction method, carbon emission prediction device, electronic apparatus, and storage medium | |
CN111967183A (en) | Method and system for calculating line loss of distribution network area | |
CN112464567B (en) | Intelligent data assimilation method based on variational and assimilative framework | |
CN112884236B (en) | Short-term load prediction method and system based on VDM decomposition and LSTM improvement | |
CN111931983A (en) | Precipitation prediction method and system | |
CN109934422A (en) | Neural network wind speed prediction method based on time series data analysis | |
CN110991660A (en) | Situation analysis method, system and storage medium of LSSVM-ARIMA model based on locust optimization | |
CN108460462A (en) | A kind of Interval neural networks learning method based on interval parameter optimization | |
CN114154401A (en) | Soil erosion modulus calculation method and system based on machine learning and observation data | |
CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
CN114578087B (en) | Wind speed uncertainty measurement method based on non-dominant sorting and stochastic simulation algorithm | |
CN114567288B (en) | Distribution collaborative nonlinear system state estimation method based on variable decibels | |
CN112231964B (en) | Gas leakage source autonomous searching and positioning method based on deep reinforcement learning | |
CN113850438A (en) | Public building energy consumption prediction method, system, equipment and medium | |
CN116993548A (en) | Incremental learning-based education training institution credit assessment method and system for LightGBM-SVM | |
CN111461327A (en) | Neural network optimization method and terminal equipment | |
CN116543259A (en) | Deep classification network noise label modeling and correcting method, system and storage medium | |
CN115310709A (en) | Power engineering project information optimization method based on particle swarm optimization | |
CN116432822A (en) | Carbon emission data prediction method, system, equipment and readable storage medium | |
CN109493065A (en) | A kind of fraudulent trading detection method of Behavior-based control incremental update |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information |
Address after: 101100 No. 1, yard 9, Hong'an street, Tongzhou District, Beijing (C2 property building, administrative office area) Applicant after: Beijing Academy of emergency management science and technology Address before: Building 4, yard 57, Yunhe East Street, Tongzhou District, Beijing 100744 Applicant before: BEIJING ACADEMY OF SAFETY SCIENCE AND TECHNOLOGY |
|
CB02 | Change of applicant information | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200410 |
|
RJ01 | Rejection of invention patent application after publication |