LSTM based prediction algorithm and abnormal change detection for temperature in aerospace gyroscope shell

Purpose–Abnormal changes in temperature directly affect the stability and reliability of a gyroscope.Predicting the temperature and detecting the abnormal change is great value for timely understanding of the working state of the gyroscope.Considering that the actual collected gyroscope shell temperature data have strong non-linearity and are accompanied by random noise pollution,the prediction accuracy and convergence speed of the traditional method need to be improved.The purpose of this paper is to use a predictive model with strong nonlinear mapping ability to predict the temperature of the gyroscope to improve the prediction accuracy and detect the abnormal change.Design/methodology/approach–In this paper,an double hiddenlayer long-shortterm memory(LSTM)is presentedto predicttemperature data for the gyroscope(including singlepoint andperiod prediction),and the evaluation index of the prediction effect is also proposed,and the prediction effects of shell temperature data are compared by BP network,support vector machine(SVM)and LSTM network.Using the estimated value detects the abnormal change of the gyroscope.Findings–By combined simulation calculation with the gyroscope measured data,the effect of different network hyperparameters on shell temperature prediction of the gyroscope is analyzed,and the LSTM networkcan beusedto predictthe temperature(timeseriesdata).Bycomparingthe performance indicatorsof different prediction methods,the accuracy of the shell temperature estimation by LSTM is better,which can meet the requirements of abnormal change detection.Quick and accurate diagnosis of different types of gyroscopefaults(stepsanddrifts)can beachievedbysettingreasonabledatawindowlengthsandthresholds.Practicalimplications–The LSTMmodelisa deepneuralnetworkmodelwithmultiplenon-linearmapping levels,and can abstract the input signal layer by layer and extract features to discover deeper underlying laws.The improved method has been used to solve the problem of strong non-linearity and random noise pollutionin time series,and the estimated value can detect the abnormal change of the gyroscope.Originality/value–In this paper,based on the LSTM network,an double hidden layer LSTM is presented to predict temperature data for the gyroscope(including single point and period prediction),and validate the effectiveness and feasibility of the algorithm by using shell temperature measurement data.The prediction effects of shell temperature data are compared by BP network,SVM and LSTM network.The LSTM network has the best prediction effect,and is used to predict the temperature of the gyroscope to improve the prediction accuracy and detect the abnormal change.

Does an abnormal change in deferred tax assets interfere with analysts' earnings forecasts?

In 2007,China adopted the single balance sheet liability method for tax accounting,but its shortcomings have emerged.I sample A-share listed companies from 2007 to 2018 to study whether an abnormal change in deferred tax assets interferes with analysts’earnings forecasts and find that an abnormal change in deferred tax assets increases the error and divergence of these forecasts.Compared with a negative abnormal change in deferred tax assets,a positive abnormal change has a greater impact on earnings forecasts.Additionally,the level of corporate governance,audit quality and analysts’professional ability have moderating effects on the correlation between an abnormal change in deferred tax assets and earnings forecasts.However,an abnormal change in deferred tax liabilities does not have a significant impact on that correlation.

Abnormal skeletal metabolism and its changes after decopper therapy in patients with Wilson disease

BACKGROUND: Researches indicate that patients with Wilson disease (WD) have abnormal skeletal metabolism, which is induced by various factors. OBJECTIVE: To probe into the changing characteristics of abnormal skeletal metabolism in WD patients and observe the effect of decopper therapy. DESIGN: Case-contrast and self-control study. SETTING: Department of Neurology, Affiliated Hospital of Neurological Institute, Anhui College of Traditional Chinese Medicine. PARTICIPANTS: A total of 35 patients with WD including 21 males and 14 females aged from 10 to 42 years with the mean age of (20±8) years were selected from Department of Neurology, Affiliated Hospital of Neurological Institute, Anhui College of Traditional Chinese Medicine from September 2000 to February 2001. All the patients were in compliance with the diagnostic criteria: history of family heredity; cone symptoms in vitro, physical sign or liver symptoms; positive Kayser-Fleischer ring; serum copper protein < 200 mg/L or A copper oxidase < 0.2; urine copper > 1.6 μmol/24 hours; liver copper > 250 μg/g (dry weight). The control group was selected from 25 cases of health individuals including 13 males and 12 females aged from 16 to 35 years with the mean age of (22±6) years. All patients who participated in the study were informed first and consented. METHODS: Patients in treatment group were treated with venous injection of 1.0 g sodium dimercaptosulfonate, once a day for totally 6 successive days. And then, patients rested for 2 days. This procedure mentioned above was regarded as a course, and the treatment lasted for 4-8 courses. Before and after injection of sodium dimercaptosulfonate, serum calcitonin (CT), osteocalcin (BGP), parathyroid hormone (PTH) and 1,25-(OH)2VitD3 were measured with radio-immunity method; blood, urine calcium, phosphorum and urine creatinine were measured with biochemical analyzer; urine dihydropyrimidine dehydrogenase(DPD) was detected with enzyme-immunity method; bone mineral density (BMD) was checked at the one third from distal end of ulna and radius with single photon absorptiometry (SPA). MAIN OUTCOME MEASURES: Relative indexes of bone metabolism of blood and urine and results of BMD in both two groups before and after treatment. RESULTS: Among 35 patients with WD and 25 healthy subjects, 5 patients were excluded because of uncompleted decopper therapy; therefore, 30 patients with WD and 25 healthy subjects were involved in the final analysis. ① Comparisons between the two groups: Contents of serum calcium, PTH and 1,25-(OH)2VitD3 were lower in treatment group than those in control group [(2.49±0.34) mmol/L vs. (2.69±0.19) mmol/L; (218.7±50.5) ng/L vs. (262.5±88.9) ng/L; (23.53±14.21) ng/L vs. (42.78±14.44) ng/L; P < 0.05-0.01]; however, contents of serum BGP and CT were higher in treatment group than those in control group [(10.22±6.11) μg/L vs. (5.78±4.22) μg/L; (282.8±109.6) ng/L vs. (62.5±37.9) ng/L, P < 0.01]; moreover, there was no significant difference of contents of serum phosphorum, urine calcium, phosphorum and DPD/creatinine between treatment group and control group (P > 0.05). BMD of males and females was lower in treatment group than that in control group [(0.617±0.197) g/cm2 vs. (0.718±0.274) g/cm2; (0.594±0.124) g/cm2 vs. (0.677±0.157) g/cm2, P < 0.05]. ② Comparisons in treatment group before and after treatment: Contents of CT and urine calcium were lower after treatment than those before treatment [(95.3±55.4) ng/L vs. (283.3±96.7) ng/L; (2.38±1.68) mmol/L vs. (3.31±2.30) mmol/L; P < 0.01, 0.05]; however, contents of 1,25-(OH)2VitD3 and DPD/creatinine were higher after treatment than those before treatment [(33.61±19.30) ng/L vs. (24.21±14.47) ng/L; (42.95±19.92) nmol/mmol vs. (19.51±9.96) nmol/mmol, P < 0.05]; moreover, there were no significant differences among other indexes before and after treatment (P > 0.05). Furthermore, there was no significant difference of BMD before and after treatment (P > 0.05). CONCLUSION: WD patients have changes in the related indexes of abnormal skeletal metabolism. In addition, contents of CT and urine calcium are decreased remarkably after decopper therapy; however, value of BMD is not changed obviously.

The Yearly Most Wind Direction Abnormality and Meteorological Disaster in Yumen in Recent 57 Years

By using the historical data during 1953-2009,the yearly most wind direction change in Yumen and the meteorological disasters of 4 times yearly most wind direction abnormality in recent 57 years were analyzed. The results showed that there were 51 years which the yearly most wind direction was the easterlies in Yumen,and the westerly had 4 years. There were 2 years which the occurrence frequencies of westerly and easterlies were same. 4 years which the yearly wind direction abnormality was the most were in 1961,1979,1987 and 1998. When the yearly wind direction abnormality was the most,the meteorological disaster was serious. The total output of grain in Gansu Province in 1961 was the least in the history in recent 60 years. The serious drought disaster in 1961 caused that half agricultural population in Gansu seriously lacked of the grain,and the dead population sharply increased. In the end of 1961,the population in Gansu decreased nearly million than in 1958. The annual precipitation in 1979 was the most in recent 57 years. The daily precipitation on June 11,1987 was the most in June of recent 57 years in Yumen. The annual average temperature in 1998 was the highest in Yumen in recent 57 years.

Analysis and Study on the Change of the Observed Geo-electric Field Data at Lhasa Geomagnetic Station before and after the Nepal MS8.1 Earthquake

In this paper,the main purpose is to analyze and research the characteristics of the geoelectric field observed data with a long time span and large amplitude abnormal change,at the Lhasa geomagnetic station( hereafter referred to as "Lhasa station "),before and after the Nepal M_S8. 1 strong earthquake,which occurred on April 25,2015. Based on the observation conditions,the observation system,and the observed data of Lhasa station preliminary discussed,the main characteristics of the abnormal change and evolution process are analyzed and studied,using the following two methods; the "synthesis energy accumulation"and the "power as MSA spectrum"analysis,from the two aspects of the"Time Domain"and"Frequency Domain. "The results show that the abnormal change of the geo-electric field observation of Lhasa station experienced a development stage following the process of "trend change- disturbance change- earthquake period-recovery period",and an evolution process of "low frequency change- high frequency change- smooth change- high frequency change ",before and after the Nepal M_S8. 1strong earthquake. Comprehensive analysis shows that the variation characteristics and evolution process of the geo-electric field at Lhasa station are basically consistent with the results of the relevant mechanism and phenomenon research. So far,this is valuable information with certain objectivity,which is typical and representative to reflect the whole process of the gestation, occurrence and complete development of such strongearthquakes.