Inrush current will lead to temperature rise of the capacitor. This temperature rise will be influenced by the max current value and the time in which it flows through the capacitor. Periodic exposure will probably increase the chance of failure of the capacitor.

Electrolytic capacitors (EC) are the essential electronic component of the power converters. Most of the methods for fault diagnosis[1] or prognosis[2] of the EC (-degradation) are, mainly, based on the parameters like : a) equivalent series resistance (ESR),   b)operating temperature (Tc), and   c)the ripple voltage (Vr).

1.   Real time remaining useful life prediction of the electrolytic capacitor https://www.semanticscholar.org/paper/Real-time-remaining-useful-life-prediction-of-the-Khandebharad-Dhumale/ffd848407b256610debe45e699a1f93a10e9a46b

2.   A prognosis case study for electrolytic capacitor degradation in DC-DC converters https://www.researchgate.net/publication/228992132_A_prognosis_case_study_for_electrolytic_capacitor_degradation_in_DC-DC_converters

And a helpful summary (A partial copy) :  Capacitor degradation is typically attributed to :

a. High Voltage conditions: The capacitance decreases and ESR increases.

b. Transients : The leakage current can be high and an internal short-circuit can occur.

c. Reverse Bias : The leakage current becomes high with loss of capacitance and increase in ESR.

d. Strong Vibrations : These can cause internal short circuits, capacitance losses, high leakage currents, high ESR and open circuits.

e. High Ripple current : These cause internal heating ,increasing the core temperature which results in gradual aging of the capacitor.

Conference Paper A prognosis case study for electrolytic capacitor degradatio...

The height of the inrush current depends to a great extend on the measurement method. In effect it is only limited by source and line resistance.

But although it may seem "extreme", the inrush current is of least concern regarding the lifetime of electrolytic capacitors as its duration is extremely short. Prof. Samaras already gave the important influencers on EC lifetime. Let me add that these result in a single parameter - the one that's the determining parameter: the internal operating temperature of the el. capacitor.

I sincerely thank Mr.Książkiewicz, Prof. Samaras and Mr. Dreher for their answers. 

As mentioned in above posts,, inrush current increases the internal operating temperature of electrolytic capacitor. I have observed the failure of a electrolytic. capacitor after 2 years of operation. Failure of capacitor led to failure of PCB and subsequent failure of entire system.Similar failure was observed on same component in different systems,  almost in same period of operation.

Suspected causes of failure was (i) high inrush current during ON/OFF of system several times a day (ii) highly humid environment. Does the combination both conditions can led to a catastrophic failure of capacitor?

@ Praveen Kesavan

Difficult to analyze such a failure remotely, especially without knowing the capacitor's specifications and operation parameters. But the "highly humid environment" is suspective: what are you talking about?

• High relative humidity  plus
• High temperature?

What would be the values? I can imagine Al corrosion due to humidity (given the caps are the the Al type).

If you have to "dig deeper":

• Get thermographic equipment to assess the cap's temperature during normal operation. Be aware that thermographic imaging affects the operating temperature if an otherwise closed containment has to be opened to enable imaging. (I consider thermometers second to thermography.)
• You might be able to install some power NTC to reduce the inrush current without compromising normal operation too much.

Given the time to fail, some experience would speed-up the locating of the root cause. Thus seeking advice from an experienced EE engineer should help significantly in finding the cause of the failures.

Thanks Mr. Dreher for your valuable suggestions.

Answering your queries, humidity is measured to be 90-95% RH and temperature is 25-28 degree Celsius. The capacitors are of Al type.

.Temperature of capacitor during operation of a fresh board was observed and there was no rise. Also fresh boards doesn't show any sign of failure during the initial phase during normal operation. The failure occurs only after a period of operation.

It is suspected that capacitor undergo continual stress, probably due to inrush current aggravated by humidity.

You could start with an optical inspection of caps on still functioning devices: I expect exposed Al to appear altered due to the formation of Al(OH)3 (which could also be substantiated by chemical analysis).

Oh, I nearly missed the "fresh board": check units that are already operating for some months. Al cap "ageing" tends to show some exponential rise in temperature (increase in ESR and decrease of capacitance value). Which could be summarized as "the older, the hotter".

The humidity[1] (measured to be 90-95% RH) is near the limits of moisture resistance under load for commercial Al-Electrolytic Capacitors[2].  Also, there is a hydrogen gas[3] out-gassing (during heating), that results in a severe and chronic mechanical stress of the encapsulation[4], upon many On/Off switching of the power.  In addition, the effects of any un-commonl[5] atmospheric stress corrosion (with high concentrations of halogen, or halogen-compound in sea water, NaCl etc.), attain it as comparable to the cautions (from corrosion) in cleaning[5] of the printed circuit boards. High corrosion from a nearby sea water (let me say, "intoxicated" humidity) is better[6] known in another (but somehow a near) topic, of Steel corrosion studies, under sea salt (NaCl etc.) aerosol proxies[6].

1. Degradation testing and failure analysis of DC film capacitors under high humidity conditions http://www.sciencedirect.com/science/article/pii/S0026271415001419

2. Polymer Aluminum Electrolytic Capacitors http://www.murata.com/~/media/webrenewal/support/library/catalog/products/capacitor/polymer/c90e.ashx?la=en-us

3. The (dielectric) oxide formation is carried out in two reaction steps

2 Al + 6 H2O → 2 Al(OH)3 + 3 H2 ↑

This reaction is accompanied by a pressure buildup in the capacitor housing caused by the released hydrogen gas. The gel-like aluminum hydroxide Al(OH)3 (ATH), is converted into aluminum oxide, Al2O3:

2 Al(OH)3 → 2 AlO(OH) + 2 H2O → Al2O3 + 3 H2O

The Al2O3 serves as dielectric and also protects the metallic aluminum against aggressive chemical reactions from the electrolyte.

4. Using Aluminum Electrolytic Capacitors http://www.elna.co.jp/en/capacitor/alumi/attention/detail.html

5. Protective housing case materials (Al, or a lower cost Al-alloys, protective housing case shell, sealed together with rubber seal). http://www.rubycon.co.jp/en/products/alumi/pdf/Process.pdf

6. a) Effect of Relative Humidity on Corrosion of Steel under Sea Salt Aerosol Proxies... https://www.researchgate.net/profile/R_Kelly/publication/274469283_Effect_of_Relative_Humidity_on_Corrosion_of_Steel_under_Sea_Salt_Aerosol_Proxies_II_MgCl2_Artificial_Seawater/links/5575e78c08ae75363751a425/Effect-of-Relative-Humidity-on-Corrosion-of-Steel-under-Sea-Salt-Aerosol-Proxies-II-MgCl2-Artificial-Seawater.pdf?origin=publication_detail&ev=pub_int_prw_xdl&msrp=LxlLwvBP7Qy0EQmPfZfYL_Dnb44dAd6eWTTHCfBVBPL1YMlGInaWUEdrTPzKrK9daXIehUtFSehDIzxnkdRO6VMUojb5abqt2ka-sUUCNxJKOQvHiq4F_MR5.6eCv99E8b6wemPWAwemLNhey7mW6U7RXaXyBw1ejx3j3AeobsjmQAtvllG1D5XgpCYWkN0hCf6VzC2gDCuOXmBy47QHA_thMBtPORw.FAmBvYksWkEU6eXEr_UZQwCIK7I_15702PaHOAlgjzOjiq0etJkNJzwjILHVo2TyHaPBnR47OKIqb4Z93kRCc2lILr0HRidQhssysw.sccfDlv1-kDvs8KFIhB-dWbTe_HIKPW0o3zlXpKtvWT2T5Hh5woc30t6ZMUH4gCwyfkEqbOQ9uuBndPfpV-U01RHz6Gxz7633nh3yA             b) 08492 Effects of Temperature and Humidity on Atmospheric Stress Corrosion Cracking of 304 Stainless Steel https://store.nace.org/aab75aef-8352-dd11-889d-0017a446694e

Article Effect of Relative Humidity on Corrosion of Steel under Sea ...

Thank you Mr. Dreher and Prof. Samaras for your valuable guidance and suggestions.

during switch on, capacitor is acting like a short circuit, that is current

Ic=v/r *e^-t/rc, so restricting this current is the only solution by using NTC in series.

With the increase in temperature, the electrolyte tends to evaporate faster. With a reduced level of electrolyte the capacitance changes and ESR increases. Now, whenever there is high inrush current it increases the temperature of the capacitor. Because of this, the life of the EC gets reduced.