CN108073134A - A kind of alarm method based on digital control system functional safety threshold value - Google Patents
A kind of alarm method based on digital control system functional safety threshold value Download PDFInfo
- Publication number
- CN108073134A CN108073134A CN201611020669.2A CN201611020669A CN108073134A CN 108073134 A CN108073134 A CN 108073134A CN 201611020669 A CN201611020669 A CN 201611020669A CN 108073134 A CN108073134 A CN 108073134A
- Authority
- CN
- China
- Prior art keywords
- control system
- numerical control
- safety threshold
- layer
- threshold value
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/406—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by monitoring or safety
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/31—From computer integrated manufacturing till monitoring
- G05B2219/31439—Alarms can be warning, alert or fault
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention relates to a kind of alarm methods based on digital control system functional safety threshold value, comprise the following steps：NC system architecture is divided into multilayer；Ask for every layer of digital control system secure threshold；When the feedback signal of digital control system is more than secure threshold alarm.The present invention is based on cnc system software architecture distribution function secure threshold indexs, can ensure the integrality and uniformity of system, while ensure that element relative weight reasonably distributes.The validity of distribution method can be improved, and foundation is provided for the feasibility of subsystem.
Description
Technical Field
The invention relates to the field of numerical control system function safety, in particular to a method for alarming according to numerical control system function safety threshold calculation.
Background
The numerical control technology is the core of advanced manufacturing technology and is the basis for realizing automation, networking, flexibility and integration in manufacturing industry. The numerical control device is used as a control center of various numerical control machines and is called as a soul and a brain of a machine tool. If the machine tool fails, abnormal actions of machine tool machining can be directly caused, so that machine tool parts or machined workpieces are damaged, and even casualties are caused. Therefore, the function safety of the numerical control device is a popular research direction of key technology in the field of numerical control at present. At present, a control system structure based on a microprocessor is mainly adopted by a numerical control device, functions such as numerical control motion control and the like are realized in a software mode of a real-time operating system and a numerical control program in the numerical control device, and generally, the numerical control motion control period is 2 milliseconds or shorter, the execution speed is high, and the execution period is short. Meanwhile, the number of safety state detection signals in the running process of the machine tool and the numerical control system is large, and the safety state detection signals simultaneously comprise hardware signals from auxiliary systems such as an overload alarm of a servo motor on the machine tool, a limit switch of the machine tool, an oil pressure alarm and an air pressure alarm, and software signals such as soft limit, out-of-tolerance and the like in the program control process of the numerical control device, the number of control signals to be monitored is large, and the correlation among the signals is complex; in addition, the real-time response speed requirement on the state judgment is high. Foreign mainstream manufacturers such as siemens germany, and genealogy japan mostly adopt full redundancy (i.e., a multiprocessor redundancy processing structure), and realize a functional security alarm by adopting a cross check method, so that the system design complexity is high, and the cost is high.
Disclosure of Invention
Aiming at the technical defects, the invention provides a method for adding a numerical control system function safety threshold value calculation device, which is used for judging the numerical control system function safety threshold value and feeding the judged value back to a numerical control system processor system, wherein system hardware comprises an ARM processor and an FPGA chip, the redundant design of a double-processor system is not required to be adopted by the numerical control system, and the cost is effectively saved while the numerical control system function safety judgment is carried out.
The technical scheme adopted by the invention for solving the technical problem is as follows: an alarm method based on a numerical control system function safety threshold comprises the following steps:
dividing a numerical control system architecture into a plurality of layers;
solving a safety threshold value of the numerical control system of each layer;
and alarming when the threshold value of each layer exceeds the set value of the numerical control system.
The multilayer includes: a numerical control system functional layer, a numerical control system program layer and a numerical control system component layer;
the numerical control system functional layer comprises a plurality of functional elements which are respectively used for realizing the communication function, the man-machine control, the decoding, the input and output, the position control, the interpolation and the parameter editing of numerical control;
the numerical control system program layer comprises a plurality of program elements, and each program element is associated with one or more functional elements and is used for dividing corresponding functions;
the numerical control system component layer comprises a plurality of component elements, and each component is associated with one or more program elements and used for representing the minimum execution unit of the numerical control system.
The step of solving the safety threshold value of the numerical control system of each layer comprises the following steps:
wherein, U F A safety threshold value of a numerical control system functional layer; omega f i Is the global relative weight of the functional element i; rf (radio frequency) wave i Is the functional safety threshold of functional element i; f =1 … n, n being a natural number.
The step of solving the safety threshold value of the numerical control system of each layer comprises the following steps:
wherein, U P The safety threshold value is a safety threshold value of a program layer of the numerical control system;ωp i is the global relative weight of the program element i; rp i Is the functional safety threshold of program element i, f =1 … n, n being a natural number.
The ω f i Is the global relative weight, ω p, of function i i The method is characterized in that the global relative weight of the program element i is obtained by an analytic hierarchy process, and comprises the following steps:
1) The current layer has m elements to be associated with the elements of the upper layer, and the weight vector of the current layer is as follows:
W＝(w 1 ,w 2 ,...,w k ) T 0≤w i ≤1(i＝1,2,...,k) (5)
2) Characteristic value w i And maximum characteristic root λ max Respectively as follows:
a is a set judgment matrix with the element of a ij ，a kj The jth element representing the kth line; (AW) i Is the ith component of the vector obtained by multiplying A by W;
3) Calculating a consistency ratio CR:
RI is a set value, and n is an order;
4) The overall rank coherency ratio for all layers is
b p Representing the set relative weight of the p-th layer relative to the architecture, wherein h is the number of layers; CI p Representing the uniformity ratio of the p-th layer; RI (Ri) p A setting value representing the p-th layer;
5) When CR' is less than or equal to 0.1, the total hierarchical ordering result satisfies consistency, and the obtained w i As global relative weight of the current level element i, i.e. ω f i Or ω p i (ii) a Otherwise, changing the value of the judgment matrix, and returning to the step 2) until CR' is less than or equal to 0.1.
The step of solving the safety threshold value of the numerical control system of each layer comprises the following steps:
wherein rp is i A safety threshold value of a numerical control system component layer; rc j For each component's functional safety threshold, c i Representing each component, j =1 … n, n being a natural number.
The invention has the following beneficial effects and advantages:
1. by adopting the method, the function safety threshold value index is distributed based on the software architecture of the numerical control system, the integrity and consistency of the system can be ensured, and the reasonable distribution of the relative weight of the elements is ensured.
2. The device is an independent monitoring device and is additionally arranged in the numerical control system, so that the design and development cost of the safe numerical control system can be effectively saved.
3. The invention can improve the effectiveness of the distribution method and provides a basis for the feasibility of the subsystem.
Drawings
FIG. 1 is a hardware configuration diagram of the apparatus of the present invention;
FIG. 2 is a flow chart of the processing steps of the method of the present invention;
FIG. 3 is a diagram of a model of a numerical control system software architecture;
FIG. 4 is a flow chart of an analytic hierarchy process;
FIG. 5 is a flow chart of a cultural algorithm;
fig. 6 is a graph of fitness as a function of iteration number.
Detailed Description
The present invention will be described in further detail with reference to examples.
The invention relates to the field of numerical control system function safety, in particular to a numerical control system function safety threshold value calculating device. The invention establishes a numerical control system architecture based on component technology. The system function safety threshold value index is distributed to the functional components, and the distribution method of the function safety threshold value is provided by taking the system practicability as an objective function and taking the reliability and the cost function of the functional components as constraint conditions. And calculating the relative weight of each layer element of the numerical control system structure by using an analytic hierarchy process, and calculating the distribution index of each component by using a cultural algorithm. By adopting the method, the functional safety threshold index of the numerical control system is ensured to meet the requirement, the development cost is effectively saved, and the effectiveness of the distribution method is improved.
The numerical control system function safety threshold calculation device is an independent monitoring device and is additionally installed in the numerical control system, system hardware comprises an ARM processor and an FPGA chip, and the purpose that the numerical control system function safety threshold is judged and fed back to a numerical control system processor system is achieved. The method comprises the steps of calculating the relative weight of each layer element of the software architecture of the numerical control system by using an analytic hierarchy process through a numerical control system software architecture model established on ARM processing, calculating the distribution value of each component by using a cultural algorithm, and calculating the safety threshold by using the reliability and cost function of the functional components of the numerical control system as constraint conditions.
Referring to fig. 1, a hardware device structure diagram suitable for the threshold of the numerical control system is shown. The device comprises an ARM processor and an FPGA chip, wherein the FPGA chip comprises a dual-port buffer memory. Hardware signals from servo motor overload alarm, machine tool limit switch, auxiliary systems such as oil pressure alarm and air pressure alarm on the machine tool are directly input into the ARM processor; software signals such as soft limit, out-of-tolerance and the like in the program control process of the numerical control device are written into the dual-port buffer memory through the bus; the ARM processor executes the threshold calculation of the numerical control system and writes back the result to the dual-port buffer memory for the numerical control system to read and execute through the bus.
According to the method, the ARM processor executes the threshold calculation of the numerical control system, and the functional safety threshold of each unit of the functional layer, the program layer and the component layer can be obtained. And writing the result into the dual-port buffer memory for the numerical control system to read and execute through the bus. The result is compared with the threshold range set by the numerical control system. Out of range, an alarm signal is generated.
Referring to fig. 2, 3 and 4, a threshold value calculation method suitable for a numerical control system includes the following steps:
firstly, a numerical control system architecture model is established. And allocating the function safety threshold indexes of all the parts of the numerical control system among all the functional components according to the established numerical control system architecture. And (4) distributing the functional safety threshold value by taking the practicability of the numerical control system as an objective function and the reliability and cost function of the functional component as constraint conditions. And calculating the relative weight of each layer element of the software architecture of the numerical control system by using an analytic hierarchy process, and finally calculating the distribution value of each component by using a cultural algorithm.
The invention is further illustrated:
the software overall structure is described according to the Gerlan & Shaw model, and is based on an open numerical control architecture. The numerical control system architecture is divided into:
(1) A functional layer of the control system, denoted by F;
(2) A numerical control system program layer, denoted by P;
(3) And the numerical control system component layer is denoted by C.
The established architecture model of the numerical control system is shown in figure 3.
The functional layer refers to communication function, man-machine control, decoding, input and output, position control, interpolation, parameter editing and other functions, the program layer is a specific division of each corresponding numerical control system functional layer, and the component is the smallest execution unit of the numerical control system.
The function safety threshold value distribution of the numerical control system software is mainly used for decomposing the function safety threshold value index of the whole system into the function safety threshold value index of the subsystem so as to guide the research and development of the subsystem. The method is mainly used for enabling the whole system to obtain a higher functional safety threshold value, so that the optimal scheme design is carried out under specific constraint conditions.
When the numerical control system software architecture is used to perform functional safety threshold value distribution, assuming that the utility of the numerical control system and the functional safety threshold value of the software function are in a linear relationship, the numerical control system safety threshold value can be expressed as:
in the formula:
ωf i is the global relative weight of function i;
rf i is the function safety threshold for function i;
ωp i is the global relative weight of program i;
rp i is the functional safety threshold for program i. f =1 … n, n being a natural number.
Suppose all components C C in the numerical control system 1 ,c 2 ,c 3 ,…c n Are independent of each other, each component is independent of the others, the program rp i The functional safety threshold of (a) is the product of the functional safety thresholds of all components:
rc j for each component's functional safety threshold, c i Representing each component, j =1 … n, n being a natural number.
According to the structural characteristics of the numerical control system, the maximum practicability of the whole numerical control system is taken as a target, the functional safety threshold index is distributed to each component, and the functional safety threshold and the cost of the component are taken as constraint conditions. Let the numerical control system have n functional components, denoted as C { C 1 ,c 2 ,c 3 ,…c n And the functional safety threshold value of the numerical control system is allocated as follows:
rc j ≤u i
rc j ≥l i
α j +q j ·rc j ≤α·v i
in the formula: functional safety threshold of component j is rc j ，0<ωp i <1，The upper limit value of the functional safety threshold of the component i is u i Lower limit value is l i ，α j Is to impose a functional safety threshold rc on the component j General cost of time, q j Representing an adjustable cost overhead. v. of i The cost value after the development of the components is finished, alpha is 1 minus the profit margin of developers, and omega is the budget cost of the software development of the numerical control system. m is less than or equal to n.
The practical influence of each layer of the numerical control system architecture on the whole system is different from each other, and the difference is represented by the weight value. In order to fairly and accurately distribute the safety threshold of the software function of the numerical control system, the weight of each layer needs to be accurately calculated. Analytic Hierarchy Process (AHP) can be used to calculate the weight of each layer element of the numerical control system architecture. The main steps of the analytic hierarchy process are shown in FIG. 4. Where the values listed in the relative importance scale proposed by Satty for the analytic hierarchy process are all integers and do not truly reflect the judgment, an improved scale calculation is applied, the relative importance of the improvement is shown in table 1.
TABLE 1 modified relative importance Scale
Assuming that k elements of the current layer are related to one index of the upper layer, the factor of the current layer is quasi-lateral to the index of the upper layer. Comparing the relative important weights to obtain the weight vector of the layer:
W＝(w 1 ,w 2 ,...,w k ) T 0≤w i ≤1(i＝1,2,...,k) (5)
calculating a characteristic value w i And maximum characteristic root λ max . The formula is as follows:
in the formula: a is a ij The importance degree ratio of the jth component to the jth component is obtained by comparing the components in pairs. a is a kj Obtaining the importance degree ratio of the kth component and the jth component by comparing the kth component and the jth component in pairs; a is a judgment matrix; (Aw) i Is the ith element of the vector Aw.
Calculating λ according to equations (6) and (7) max And W, the weight of the factor of the hierarchy to the factor of the upper layer can be obtained. For this hierarchical single-rank computation, the consistency of the full value is such that to rank according to the criteriaIs very necessary.
Expressing the global relative weight of a functional layer of the numerical control system by using WF; and WP represents the global relative weight of the program layer of the numerical control system, and WCi represents the global relative weight of the component layer of the numerical control system. It can be found that:
wp i represents the local relative weights of the program layer of the numerical control system, whereby the above equation can be converted to:
also, the WC:
WC i local relative weights representing the component layers of the numerical control system, whereby the above equation can be converted to:
in order to check the consistency between the importance of each element, the consistency of the matrix is confirmed and discriminated by a CR (consistency ratio) value, and the consistency is satisfied when CR is less than or equal to 0.1. If the value range does not conform, the value of the judgment matrix element should be changed. The formula for CR is public (12):
in the formula: CR: judging a matrix consistency index; RI: averaging random consistency indexes; lambda [ alpha ] max : a maximum feature root; n: the order of the steps.
When consistency is checked, according to RI value, as shown in Table 2;
TABLE 2RI values
Finally, the sequence of all the layers is obtained, and the total sequence consistency ratio is as follows:
in the formula: b p Is the relative weight of the p-th layer relative to the architecture; CI p Representing the uniformity ratio of the p-th layer; RI (Ri) p The setting value of the p-th layer is shown. h is the number of layers.
When CR' is less than or equal to 0.1, the overall hierarchical ordering result is considered to have satisfactory consistency. If the value range is not in accordance, the value of the judgment matrix element should be changed.
The cultural algorithm flow chart is shown in fig. 5.
The method comprises the following steps of solving a functional safety threshold value distribution model of the numerical control system by a culture algorithm as follows:
(1) Initializing a population space: 40 individuals were randomly generated, the variable in each individual { rc 1 ,rc 2 ,rc 3 ,rc 4 ,rc 5 ,rc 6 Are generated according to the constraint in equation (4)
The encoding mode of the individual in the population space can be expressed as formula (15)
X＝[x 1 ,…x n ] (15)
(2) Setting a fitness function: the greater the practicability of the numerical control system is, the more the obtained function safety threshold index of each functional component can meet the requirement, so the fitness function is set as shown in the formula (16), and p is less than or equal to n.
(3) Setting a belief space: and setting a belief space according to the set initial space and the belief space structure. Belief space adoption<S,N>, structure pair, where S is structure knowledge, i.e. S = { S = { (S) } 1 ,s 2 L,s n Taking the maximum fitness as an initial value of the structure knowledge; n is specification knowledge, i.e. N = { B 1 ,B 2 L,B n And initializing the specification knowledge. Canonical knowledge is used to describe the feasible solution space for the problem. The updating of canonical knowledge is embodied as a change in the feasible search space.
(4) And carrying out mutation operation on the individuals in the population space according to the influence function, and generating corresponding individuals. The main role of the influence function is to use various types of knowledge guidance in the confidence space.
(5) Setting a receiving function: the individuals that can influence the belief space are selected according to the receive function. The receiving function selects better individuals from the population space, and submits the better individuals to the credibility space for research core of the receiving function in selecting the number of the better individuals.
(4) The step (5) is realized specifically as follows: and generating a child individual according to the existing individual (adjusting the variable change step length by using the canonical knowledge and adjusting the change direction by using the structural knowledge). And placing the generated offspring individuals behind the parent individuals, sequencing all the individuals according to the fitness, and placing the fitness in front. The first 40 best individuals were taken as new parents. Based on the first 40 individuals, the structural knowledge and the normative knowledge are updated with a certain probability (set to 0.7) based on the fitness.
(6) And when the set fitness function value error meets the set error requirement, stopping searching. And outputting each index corresponding to the optimal solution, namely the distribution index of the numerical control system software function safety threshold to be solved. The optimal solution is the individual with the maximum fitness in all the individuals.
TABLE 3
The maximum fitness result after the first 8 iterations is shown in table 3. The curve of the variation of the iteration times along with the evolution times is shown in fig. 6, the abscissa is the iteration times, and the ordinate is the maximum fitness. The practical index is the largest at this time. The result shows that the method ensures that the safety threshold index of the numerical control system meets the requirement, effectively saves the development cost, improves the effectiveness of the distribution method, and provides a basis for the feasibility research of the subsystem.
Claims (6)
1. An alarm method based on a numerical control system function safety threshold is characterized by comprising the following steps:
dividing a numerical control system architecture into a plurality of layers;
solving the safety threshold of the numerical control system of each layer;
and alarming when the threshold value of each layer exceeds the set value of the numerical control system.
2. The alarm method based on the numerical control system function safety threshold value as claimed in claim 1, wherein the multiple layers comprise: a numerical control system functional layer, a numerical control system program layer and a numerical control system component layer;
the numerical control system functional layer comprises a plurality of functional elements which are respectively used for realizing the communication function, the man-machine control, the decoding, the input and output, the position control, the interpolation and the parameter editing of numerical control;
the numerical control system program layer comprises a plurality of program elements, and each program element is associated with one or more functional elements and is used for dividing corresponding functions;
the numerical control system component layer comprises a plurality of component elements, and each component is associated with one or more program elements and used for representing the minimum execution unit of the numerical control system.
3. The alarm method based on the numerical control system function safety threshold value of claim 1, wherein the step of solving the numerical control system safety threshold value of each layer comprises the following steps:
wherein, U F A safety threshold value of a numerical control system functional layer; omega f i Is the global relative weight of the functional element i; rf (radio frequency) i Is the functional safety threshold of functional element i; f =1 … n, n being a natural number.
4. The alarm method based on numerical control system function safety threshold value of claim 1, wherein the finding the numerical control system safety threshold value of each layer comprises:
wherein, U P The safety threshold value is a safety threshold value of a program layer of the numerical control system; ω p i Is the global relative weight of the program element i; rp i Is the functional safety threshold of program element i, f =1 … n, n being a natural number.
5. The alarm method based on the numerical control system function safety threshold as claimed in claim 3, wherein ω f i Is the global relative weight, ω p, of function i i The method is characterized in that the global relative weight of the program element i is obtained by an analytic hierarchy process, and comprises the following steps:
1) The current layer has m elements associated with the elements of the upper layer, and the weight vector of the current layer is as follows:
W＝(w 1 ，w 2 ，...，w k ) T 0≤w i ≤1(i＝1,2,...,k) (5)
2) Characteristic value w i And maximum characteristic root λ max Respectively as follows:
a is a set judgment matrix with the element of a ij ，a kj The jth element representing the kth line; (AW) i Is the ith component of the vector obtained by multiplying A by W;
3) Calculating a consistency ratio CR:
RI is a set value, and n is an order;
4) The overall rank coherency ratio for all layers is
b p Representing the set relative weight of the p-th layer relative to the architecture, wherein h is the number of layers; CI p Representing the uniformity ratio of the p-th layer; RI (Ri) p A setting value representing the p-th layer;
5) When CR' is less than or equal to 0.1, the total hierarchical ordering result satisfies consistency, and the obtained w i As global relative weight of the current level element i, i.e. ω f i Or ω p i (ii) a Otherwise, changing the value of the judgment matrix, and returning to the step 2) until CR' is less than or equal to 0.1.
6. The alarm method based on the numerical control system function safety threshold value of claim 1, wherein the step of solving the numerical control system safety threshold value of each layer comprises the following steps:
wherein rp is i The safety threshold value of the numerical control system component layer is set; rc j For each component's functional safety threshold, c i Representing each component, j =1 … n, n being a natural number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611020669.2A CN108073134A (en) | 2016-11-18 | 2016-11-18 | A kind of alarm method based on digital control system functional safety threshold value |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611020669.2A CN108073134A (en) | 2016-11-18 | 2016-11-18 | A kind of alarm method based on digital control system functional safety threshold value |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108073134A true CN108073134A (en) | 2018-05-25 |
Family
ID=62160508
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611020669.2A Pending CN108073134A (en) | 2016-11-18 | 2016-11-18 | A kind of alarm method based on digital control system functional safety threshold value |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108073134A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2187871B (en) * | 1986-03-13 | 1989-11-15 | Shorrock Security Systems Ltd | Positional sensor |
| CN104408326A (en) * | 2014-12-16 | 2015-03-11 | 电子科技大学 | Assessment method for deep space exploration autonomous-navigation filtering algorithm |
| CN105677759A (en) * | 2015-12-30 | 2016-06-15 | 国家电网公司 | Alarm correlation analysis method in communication network |
-
2016
- 2016-11-18 CN CN201611020669.2A patent/CN108073134A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2187871B (en) * | 1986-03-13 | 1989-11-15 | Shorrock Security Systems Ltd | Positional sensor |
| CN104408326A (en) * | 2014-12-16 | 2015-03-11 | 电子科技大学 | Assessment method for deep space exploration autonomous-navigation filtering algorithm |
| CN105677759A (en) * | 2015-12-30 | 2016-06-15 | 国家电网公司 | Alarm correlation analysis method in communication network |
Non-Patent Citations (1)
| Title |
|---|
| 谷岩: "数控系统软件可靠性设计与故障分析技术", 《中国博士学位论文全文数据库 工程科技I辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109934337B (en) | Spacecraft telemetry data anomaly detection method based on integrated LSTM | |
| US20210041862A1 (en) | Malfunction early-warning method for production logistics delivery equipment | |
| CN110943983B (en) | Network security prevention method based on security situation awareness and risk assessment | |
| CN110686633B (en) | Landslide displacement prediction method and device and electronic equipment | |
| Shen et al. | Fault diagnosis and fault-tolerant control based on adaptive control approach | |
| CN104597842A (en) | BP neutral network heavy machine tool thermal error modeling method optimized through genetic algorithm | |
| CN113298297B (en) | Wind power output power prediction method based on isolated forest and WGAN network | |
| Song et al. | A stochastic approach for the reliability evaluation of multi-state systems with dependent components | |
| CN112183835B (en) | Water guide tile temperature trend early warning method, device and system based on machine learning | |
| Hwang et al. | Detection and correction method of erroneous data using quantile pattern and LSTM | |
| CN112421631A (en) | New energy consumption capacity assessment method and system | |
| Sharifi et al. | Reliability optimization of a k-out-of-n series-parallel system with warm standby components | |
| CN109918280B (en) | Health management method and device for servo system | |
| Ye et al. | Integrated design of fault detection systems in time-frequency domain | |
| Zhang et al. | Research on tool wear prediction based on LSTM and ARIMA | |
| Yang et al. | Reliability modelling of CNC machine tools based on the improved maximum likelihood estimation method | |
| CN108073134A (en) | A kind of alarm method based on digital control system functional safety threshold value | |
| Suhail et al. | Cutting parameters identification using multi adaptive network based Fuzzy inference system: An artificial intelligence approach | |
| CN111108738A (en) | Data processing device, data analysis device, data processing system, and method for processing data | |
| Roy | Modelling and predicting surface finish in turning using an adaptive neuro-fuzzy inference system | |
| Ngoc Phien et al. | A comparison between deep belief network and LSTM in chaotic time series forecasting | |
| Yan et al. | A Data Cleaning Framework for Water Quality Based on NLDIW-PSO Based Optimal SVR | |
| CN115834424B (en) | Identification and correction method for abnormal data of power distribution network line loss | |
| CN112653129B (en) | Transient power angle stability margin estimation method, device and system | |
| Kim et al. | A new approach for robust and reduced order fault detection filter design |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180525 |
|
| RJ01 | Rejection of invention patent application after publication |